On Thu, Nov 17, 2016 at 05:40:35PM +0000, Sudeep Holla wrote: > > > On 17/11/16 17:31, Thierry Reding wrote: > > On Thu, Nov 17, 2016 at 05:28:53PM +0000, Sudeep Holla wrote: > > > > > > > > > On 17/11/16 17:16, Thierry Reding wrote: > > > > From: Thierry Reding <treding@xxxxxxxxxx> > > > > > > > > The power management controller on Tegra186 has changed in backwards- > > > > incompatible ways with respect to earlier generations. This implements a > > > > new driver that supports inversion of the PMU interrupt as well as the > > > > "recovery", "bootloader" and "forced-recovery" reboot commands. > > > > > > > > Signed-off-by: Thierry Reding <treding@xxxxxxxxxx> > > > > --- > > > > .../bindings/arm/tegra/nvidia,tegra186-pmc.txt | 34 +++++ > > > > drivers/soc/tegra/Makefile | 2 +- > > > > drivers/soc/tegra/pmc-tegra186.c | 169 +++++++++++++++++++++ > > > > 3 files changed, 204 insertions(+), 1 deletion(-) > > > > create mode 100644 Documentation/devicetree/bindings/arm/tegra/nvidia,tegra186-pmc.txt > > > > create mode 100644 drivers/soc/tegra/pmc-tegra186.c > > > > > > > > > > [...] > > > > > > > diff --git a/drivers/soc/tegra/pmc-tegra186.c b/drivers/soc/tegra/pmc-tegra186.c > > > > new file mode 100644 > > > > index 000000000000..ee28eddd8e3c > > > > --- /dev/null > > > > +++ b/drivers/soc/tegra/pmc-tegra186.c > > > > > > [...] > > > > > > > + > > > > + /* > > > > + * If available, call the system restart implementation that was > > > > + * registered earlier (typically PSCI). > > > > + */ > > > > + if (pmc->system_restart) { > > > > + pmc->system_restart(reboot_mode, cmd); > > > > + return NOTIFY_DONE; > > > > + } > > > > + > > > > > > IIUC, Tegra186 implements PSCI v1.0 and it always takes above path. > > > So what other platforms does this driver support ? The name is > > > pmc-tegra186.c, hence the confusion. > > > > It's technically possible to run Tegra186 without PSCI enabled, or even > > boot it with firmware that doesn't implement PSCI. In such cases it > > OK, with enable-method as "spin-table" I suppose ? Yes, that would probably be the other choice for SMP. > > would be nice to still be able to reboot via the PMC code above. > > I assume it's only for development purposes as you won't have much CPU > power management support without PSCI. This is what I had primarily in mind. Sometimes it might be useful to boot without PSCI at all (as a poor man's way of disable SMP), in which case this driver can still be used for reboot into the bootloader or recovery mode. Thierry
Attachment:
signature.asc
Description: PGP signature