YesïINIT would make sense for me.- *_INIT and *GETTIMEOUT. Although your driver does not need them, could you take them as options in your driver ?The driver already has SMCWD_INFO which is used during probe to
retrieve the minimum and maximum timeout values supported by the
hardware at probe time. Maybe it would make sense to rename that to
INIT (which would still return those values, but can also do whatever
initialization needs to be done in TF)?
GETTIMELEFT I agree we canI just wanted it to be compatible with other OSïand IÂ think it over, maybe I can also use the
implement optionally, and other platforms would just return a
PSCI_RET_NOT_SUPPORTED for that.
- *_RESETNOW. It is used to reset the system right now, similar to your SOFT RESET. could you reserve an operation index in ATF ?Just curious, why do you need this? Shouldn't you use the PSCI
standard SYSTEM_RESET SMC for that? (If you want to control exactly
how the platform is reset, you could also use SYSTEM_RESET2 with a
vendor-defined reset_type.)
.