On Tuesday 12 April 2016 11:33 PM, Jon Hunter wrote:
On 12/04/16 17:59, Laxman Dewangan wrote:
On Tuesday 12 April 2016 08:58 PM, Thierry Reding wrote:
* PGP Signed by an unknown key
On Tue, Apr 12, 2016 at 08:26:42PM +0530, Laxman Dewangan wrote:
+#define TEGRA_IO_RAIL_EMMC 35
#define TEGRA_IO_RAIL_CAM 36
#define TEGRA_IO_RAIL_RES 37
+#define TEGRA_IO_RAIL_EMMC2 37
We have a duplicate entry for 37 now. The _RES might have meant
"reserved", in which case maybe just replace it with the new symbolic
name?
OK, then make sense to replace RES with EMMC2.
Looking at the Tegra124 TRM it was reserved and so renaming makes sense
here. However, that also prompts the question how do we check to ensure
that the IO rail is valid for a given SoC?
Should we define a 'valid' mask for IO_DPD_STATUS and IO_DPD2_STATUS
registers in the SoC data so we can check if the rail is valid?
Yes, that is good idea.
Infact, we should decouple RAIL_ID with the bit location of register.
This will help on mapping any rail ID to SoC specific bit location and
need not to worry if bit location of rail get changed on any generation.
Local lookup table from ID to bit location can make validation as well
as the decoupling.
--
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
