Re: [PATCH 5/7] PPS: serial clients support.

[Rodolfo Giometti]

On Thu, Mar 20, 2008 at 01:04:00PM -0700, Andrew Morton wrote:
> On Thu,  6 Mar 2008 13:09:04 +0100
> Rodolfo Giometti <giometti@xxxxxxxx> wrote:
> 
> > Adds support for the PPS sources connected with the CD (Carrier
> > Detect) pin of a serial port.
> > 
> > ...
> >
> > +#ifdef CONFIG_PPS_CLIENT_UART
> > +
> > +static int
> > +uart_register_pps_port(struct uart_state *state, struct uart_port *port)
> > +{
> > +	struct tty_driver *drv = port->info->tty->driver;
> > +	int ret;
> > +
> > +	state->pps_info.owner = THIS_MODULE;
> > +	state->pps_info.dev = port->dev;
> > +	snprintf(state->pps_info.name, PPS_MAX_NAME_LEN, "%s%d",
> > +		drv->driver_name, port->line);
> > +	snprintf(state->pps_info.path, PPS_MAX_NAME_LEN, "/dev/%s%d",
> > +		drv->name, port->line);
> 
> umm, why are we hard-wiring "/dev" into the kernel source?  Is it just for
> user-friendly printks?

Yes... :)

> > +	state->pps_info.mode = PPS_CAPTUREBOTH | \
> > +			PPS_OFFSETASSERT | PPS_OFFSETCLEAR | \
> > +			PPS_CANWAIT | PPS_TSFMT_TSPEC;
> > +
> > +	ret = pps_register_source(&state->pps_info, PPS_CAPTUREBOTH | \
> > +				PPS_OFFSETASSERT | PPS_OFFSETCLEAR);
> > +	if (ret < 0) {
> > +		dev_err(port->dev, "cannot register PPS source \"%s\"\n",
> > +						state->pps_info.path);
> > +		return ret;
> > +	}
> > +	port->pps_source = ret;
> > +	dev_dbg(port->dev, "PPS source #%d \"%s\" added\n",
> > +		port->pps_source, state->pps_info.path);
> > +
> > +	return 0;
> > +}
> > +
> > +static void
> > +uart_unregister_pps_port(struct uart_state *state, struct uart_port *port)
> > +{
> > +	pps_unregister_source(port->pps_source);
> > +	dev_dbg(port->dev, "PPS source #%d \"%s\" removed\n",
> > +				port->pps_source, state->pps_info.path);
> > +}
> > +
> > +#else
> > +
> > +#define uart_register_pps_port(state, port)	do { } while (0)
> > +#define uart_unregister_pps_port(state, port)	do { } while (0)
> 
> Please never implement in cpp that which can be implemented in C.  These
> should have been static inlines.  That's nicer, matches the
> CONFIG_PPS_CLIENT_UART=y code and provides type checking.

Are you saying that I should do:

static inline int
uart_register_pps_port(struct uart_state *state, struct uart_port *port)
{
	/* Nop */
}

static inline void
uart_unregister_pps_port(struct uart_state *state, struct uart_port *port)
{
	/* Nop */
}

?

Ciao,

Rodolfo

-- 

GNU/Linux Solutions                  e-mail:    giometti@xxxxxxxxxxxx
Linux Device Driver                             giometti@xxxxxxxxx
Embedded Systems                     		giometti@xxxxxxxx
UNIX programming                     phone:     +39 349 2432127
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/