On Wed, May 06, 2020 at 05:44:09PM +0100, Sudeep Holla wrote: > In order to add newer SMCCC v1.1+ functionality and to avoid cluttering > PSCI firmware driver with SMCCC bits, let us move the SMCCC specific > details under drivers/firmware/smccc/smccc.c > > We can also drop conduit and smccc_version from psci_operations structure > as SMCCC was the sole user and now it maintains those. > > No functionality change in this patch though. > > Signed-off-by: Sudeep Holla <[email protected]> > --- > MAINTAINERS | 9 +++++++++ > drivers/firmware/Makefile | 3 ++- > drivers/firmware/psci/psci.c | 19 ++++--------------- > drivers/firmware/smccc/Makefile | 3 +++ > drivers/firmware/smccc/smccc.c | 26 ++++++++++++++++++++++++++ > include/linux/arm-smccc.h | 11 +++++++++++ > include/linux/psci.h | 2 -- > 7 files changed, 55 insertions(+), 18 deletions(-) > create mode 100644 drivers/firmware/smccc/Makefile > create mode 100644 drivers/firmware/smccc/smccc.c > > Hi Mark, Lorenzo, > > I have replicated PSCI entry in MAINTAINERS file and added myself to > for SMCCC entry. If you prefer I can merge it under PSCI. Let me know > your preference along with other review comments.
> +SECURE MONITOR CALL(SMC) CALLING CONVENTION (SMCCC) > +M: Mark Rutland <[email protected]> > +M: Lorenzo Pieralisi <[email protected]> > +M: Sudeep Holla <[email protected]> > +L: [email protected] > +S: Maintained > +F: drivers/firmware/smccc/ > +F: include/linux/arm-smccc.h As per the above, I'm fine with having this separate from the PSCI entry, and I'm fine with sharing this maintainership. > diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h > +/** > + * arm_smccc_version_init() - Sets SMCCC version and conduit > + * @version: SMCCC version v1.1 or above > + * @conduit: SMCCC_CONDUIT_SMC or SMCCC_CONDUIT_HVC > + * > + * When SMCCCv1.1 or above is not present, defaults to ARM_SMCCC_VERSION_1_0 > + * and SMCCC_CONDUIT_NONE respectively. > + */ > +void __init arm_smccc_version_init(u32 version, enum arm_smccc_conduit > conduit); Given we only expect the PSCI code to call this, could we avoid putting this in a header and just define it within psci.c? Either way: Acked-by: Mark Rutland <[email protected]> Thanks, Mark.
