> -----Original Message-----
> From: Arce, Abraham
> Sent: Thursday, May 06, 2010 9:49 AM
> To: G, Manjunath Kondaiah; Tony Lindgren
> Cc: linux-omap@xxxxxxxxxxxxxxx;
> spi-devel-general@xxxxxxxxxxxxxxxxxxxxx
> Subject: RE: [PATCH v1 2/3] OMAP4: Ethernet: KS8851 Board Support
>
> Manjunath,
>
> > > > > > > > +
> > > > > > > > +static void omap_ethernet_init(void)
> > > > > > > > +{
> > > > > > > > +
> gpio_request(ETHERNET_KS8851_POWER_ENABLE, "ethernet");
> > > > > > > > +
> gpio_direction_output(ETHERNET_KS8851_POWER_ENABLE, 1);
> > > > > > > > + gpio_request(ETHERNET_KS8851_QUART, "quart");
> > > > > > > > + gpio_direction_output(ETHERNET_KS8851_QUART, 1);
> > > > > > > > + gpio_request(ETHERNET_KS8851_IRQ, "ks8851");
> > > > > > >
> > > > > > > Any reason for not checking return value of
> gpio_request()?
> > > > > > >
>
> [snip]
>
> > Minor changes:
> > if (gpio_request(x))
> > return status;
> > if (gpio_request(y))
> > goto err1;
> > if (gpio_request(z))
> > goto err2;
> > ...
> > return 0;
> >
> > err2:
> > free(y);
> >
> > err1:
> > free(x);
> > return status;
>
> How about this patch?
>
> Note. I am thinking in spi_register_board_info registration:
>
> 1. to be done if ethernet fails, other spi dev can be added in future
No use. Since, for a given board SPI chip select lines are physically
connected to external slave devices, you can't register another SPI
slave device for same chip select.
> 2. not to be done if ethernet fails, no other dev in spi bus 1 for now
>
> What is the best approach? Is it to keep #2?
Seems to be ok for chip select #0. If we have another device on bus 1 with
different chip select number, then registration should succeed. However, this
may not be the case with omap4 sdp board since it might have only one spi
device on bus1.
-Manjunath--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
