Re: [PATCH v3 -next 2/3] ACPI: RISC-V: Add LPI driver

[Sunil V L]

On Sun, Mar 17, 2024 at 10:33:15PM -0700, Drew Fustini wrote:
> On Thu, Jan 18, 2024 at 11:59:29AM +0530, Sunil V L wrote:
> > Enable Low Power Idle (LPI) based cpuidle driver for RISC-V platforms.
> > It depends on SBI HSM calls for idle state transitions.
> > 
> > Signed-off-by: Sunil V L <sunilvl@xxxxxxxxxxxxxxxx>
> > Reviewed-by: Andrew Jones <ajones@xxxxxxxxxxxxxxxx>
> > ---
> >  drivers/acpi/riscv/Makefile  |  3 +-
> >  drivers/acpi/riscv/cpuidle.c | 81 ++++++++++++++++++++++++++++++++++++
> >  2 files changed, 83 insertions(+), 1 deletion(-)
> >  create mode 100644 drivers/acpi/riscv/cpuidle.c
> > 
> > diff --git a/drivers/acpi/riscv/Makefile b/drivers/acpi/riscv/Makefile
> > index 8b3b126e0b94..7309d92dd477 100644
> > --- a/drivers/acpi/riscv/Makefile
> > +++ b/drivers/acpi/riscv/Makefile
> > @@ -1,2 +1,3 @@
> >  # SPDX-License-Identifier: GPL-2.0-only
> > -obj-y 	+= rhct.o
> > +obj-y					+= rhct.o
> > +obj-$(CONFIG_ACPI_PROCESSOR_IDLE)	+= cpuidle.o
> > diff --git a/drivers/acpi/riscv/cpuidle.c b/drivers/acpi/riscv/cpuidle.c
> > new file mode 100644
> > index 000000000000..624f9bbdb58c
> > --- /dev/null
> > +++ b/drivers/acpi/riscv/cpuidle.c
> > @@ -0,0 +1,81 @@
> > +// SPDX-License-Identifier: GPL-2.0-only
> > +/*
> > + * Copyright (C) 2024, Ventana Micro Systems Inc
> > + *	Author: Sunil V L <sunilvl@xxxxxxxxxxxxxxxx>
> > + *
> > + */
> > +
> > +#include <linux/acpi.h>
> > +#include <acpi/processor.h>
> > +#include <linux/cpu_pm.h>
> > +#include <linux/cpuidle.h>
> > +#include <linux/suspend.h>
> > +#include <asm/cpuidle.h>
> > +#include <asm/sbi.h>
> > +#include <asm/suspend.h>
> > +
> > +#define RISCV_FFH_LPI_TYPE_MASK	GENMASK_ULL(63, 60)
> > +#define RISCV_FFH_LPI_RSVD_MASK	GENMASK_ULL(59, 32)
> > +
> > +#define RISCV_FFH_LPI_TYPE_SBI	BIT_ULL(60)
> > +
> > +static int acpi_cpu_init_idle(unsigned int cpu)
> > +{
> > +	int i;
> > +	struct acpi_lpi_state *lpi;
> > +	struct acpi_processor *pr = per_cpu(processors, cpu);
> > +
> > +	if (unlikely(!pr || !pr->flags.has_lpi))
> > +		return -EINVAL;
> > +
> > +	if (!riscv_sbi_hsm_is_supported())
> > +		return -ENODEV;
> > +
> > +	if (pr->power.count <= 1)
> > +		return -ENODEV;
> > +
> > +	for (i = 1; i < pr->power.count; i++) {
> > +		u32 state;
> > +
> > +		lpi = &pr->power.lpi_states[i];
> > +
> > +		/*
> > +		 * Validate Entry Method as per FFH spec.
> > +		 * bits[63:60] should be 0x1
> > +		 * bits[59:32] should be 0x0
> > +		 * bits[31:0] represent a SBI power_state
> > +		 */
> > +		if (((lpi->address & RISCV_FFH_LPI_TYPE_MASK) != RISCV_FFH_LPI_TYPE_SBI) ||
> > +		    (lpi->address & RISCV_FFH_LPI_RSVD_MASK)) {
> > +			pr_warn("Invalid LPI entry method %#llx\n", lpi->address);
> > +			return -EINVAL;
> > +		}
> > +
> > +		state = lpi->address;
> 
> It seems that acpi_lpi_state.address is u64, so shouldn't state be u64
> instead of u32?
> 
SBI suspend state is only 32 bits represented by lower 32 bits of
lpi->address.

Thanks,
Sunil