On Wed, Jun 23, 2010 at 00:03:47 +0200, Linus Walleij wrote:
> 2010/6/22 Rabin Vincent <rabin.vincent@xxxxxxxxxxxxxx>:
> > @@ -17,7 +17,8 @@
> > * mmc/host.h
> > * @translate_vdd: a callback function to translate a MMC_VDD_*
> > * mask into a value to be binary or:ed and written into the
> > - * MMCIPWR register of the block
> > + * MMCIPWR register of the block. May also control external power
> > + * based on the power_mode.
>
> Actually this callback is named like that for a reason: it is to be used
> to translate the MMC_VDD_* into a u8:4 bitmask for the MMCIPWR
> register, 4 bits possibly routed out of the PL180 block on some
> designs (never seen in practice, but could be used in theory).
>
> So it translates a voltage into a 4-bit bitmask, hence the name.
>
> Now the semantics are altered to have other side-effects in
> the platform, so the function should be renamed, something like
> platform_vdd_handler() would be more appropriate.
>
This is already inside platform data, so how about just ->vdd_handler()?
Or ->set_power(), like some other drivers?
Note that we'd also like to use this to do the board-specific settings
for the *DIREN and FBCLK bits, which replace the Voltage bits on some ST
variants, like so:
+static u32 mop500_sdi0_translate_vdd(struct device *dev, unsigned int vdd,
+ unsigned char power_mode)
+{
+ if (power_mode == MMC_POWER_UP)
+ gpio_set_value(gpio_sdmmc_en, 1);
+ else if (power_mode == MMC_POWER_OFF)
+ gpio_set_value(gpio_sdmmc_en, 0);
+
+ return MCI_FBCLKEN | MCI_CMDDIREN | MCI_DATA0DIREN |
+ MCI_DATA2DIREN | MCI_DATA31DIREN;
+}
Rabin
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
