On Wed, Oct 09, 2013 at 03:09:21PM +0200, Linus Walleij wrote:
> When I look at the i.MX driver (I guess this is what Shawn
> is working on here) it can actually be augmented to do it
> the same way and avoid this mess altogether:
>
> for (i = 0; i < num_configs; i++) {
> if (info->flags & SHARE_MUX_CONF_REG) {
> u32 reg;
> reg = readl(ipctl->base + pin_reg->conf_reg);
> reg &= ~0xffff;
> reg |= configs[i];
> writel(reg, ipctl->base + pin_reg->conf_reg);
> } else {
> writel(configs[i], ipctl->base + pin_reg->conf_reg);
> }
> dev_dbg(ipctl->dev, "write: offset 0x%x val 0x%lx\n",
> pin_reg->conf_reg, configs[i]);
> } /* for each config */
>
> As can be seen it will just writel() the config into the register
> for each config, it seems more like there is a bug that also
> the else-clause here should be read-modify-write, or am
> I getting it all wrong?
On imx, num_configs is always 1 (new_map[j].data.configs.num_configs
is hard-coded as 1 in imx_dt_node_to_map() function). We have all
configs for given pin encoded in the last integer of fsl,pin entry.
And this integer can be directly written into conf_reg.
Shawn
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
