On Thu, Jun 20, 2013 at 12:59 PM, Michal Simek <mon...@monstr.eu> wrote: > On 06/20/2013 11:23 AM, Linus Walleij wrote:
>> What about something like this: >> >> static bool is_dual (struct device_node *np) >> { >> struct property *prop = of_find_property(np, "xlnx,is-dual", NULL); >> int ret; >> u32 val; >> >> if (!prop) >> return false; >> >> ret = of_property_read_u32(np, "xlnx,is-dual", &val); >> if (ret < 0) >> return true; /* node exists but has no cells */ >> >> return !!val; >> } > > we can do it in this way but what I don't like on this is that IP > is design to support 2 channels right now. > It can happen that Xilinx decides to extend this for new channels. > Register map is prepared for it and there is enough space to do it. > > And when this is done then is-dual (which is current name which > is used in hardware configuration from design tools) will contain > larger value >1. > I agree that is-dual is not the best name. You don't say. It's about as smart as this: #define NUMBER_TWO 4 Oh well, that's not your fault. But seriously: xlnx,is-dual; without an argument can still be taken to mean "2", and you still need to inperpret the absence if this parameter as "1" do you not? > What about to do it differently? > Generate number of channel in the description. > And also do for() loop in the probe function to read values based > on this channel number. Sorry I can't translate this, can you send a patch? Yours, Linus Walleij -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/