On Thu, Aug 05, 2010 at 08:14:55AM -0700, Tabi Timur-B04825 wrote:
Mark Brown wrote:
This seems crazy, it means that we're not able to use new support for hardware features to the driver which require any kind of flag or data without also going through and updating the device trees for all existing boards. That doesn't seem terribly helpful.
Hey, what do you want me to do? That's just the way the hardware is designed. It is not possible for software to know the FIFO depth of the SSI on its own, because there is no register that can be queried. The device tree was designed specifically for this purpose -- the provide a data structure (e.g. not kernel code) that describes all of the unprobe-able features of the various devices on the SOC.
This doesn't seem like a hardware issue, it seems like an issue with the way we've deployed the device tree. I'd have strongly expected that the device tree was able to incoporate all the properties that are standard to the CPU by reference somehow (with that data being distributed separately to the per system device tree).