freebsd-dev/sys/dev/acpica/acpi_cpu.c
John Baldwin fc0ea94a7d Catch up to changes in acpivar.h to add support for using ACPI on
4-stable systems.

Sponsored by:	The Weather Channel
2002-10-16 17:28:53 +00:00

415 lines
12 KiB
C

/*-
* Copyright (c) 2001 Michael Smith
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
#include "opt_acpi.h"
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/bus.h>
#include <sys/power.h>
#include <machine/bus_pio.h>
#include <machine/bus.h>
#include <machine/resource.h>
#include <sys/rman.h>
#include "acpi.h"
#include <dev/acpica/acpivar.h>
/*
* Support for ACPI Processor devices.
*
* Note that this only provides ACPI 1.0 support (with the exception of the
* PSTATE_CNT field). 2.0 support will involve implementing _PTC, _PCT,
* _PSS and _PPC.
*/
/*
* Hooks for the ACPI CA debugging infrastructure
*/
#define _COMPONENT ACPI_PROCESSOR
ACPI_MODULE_NAME("PROCESSOR")
struct acpi_cpu_softc {
device_t cpu_dev;
ACPI_HANDLE cpu_handle;
u_int32_t cpu_id;
/* CPU throttling control register */
struct resource *cpu_p_blk;
#define CPU_GET_P_CNT(sc) (bus_space_read_4(rman_get_bustag((sc)->cpu_p_blk), \
rman_get_bushandle((sc)->cpu_p_blk), \
0))
#define CPU_SET_P_CNT(sc, val) (bus_space_write_4(rman_get_bustag((sc)->cpu_p_blk), \
rman_get_bushandle((sc)->cpu_p_blk), \
0, (val)))
#define CPU_P_CNT_THT_EN (1<<4)
};
/*
* Speeds are stored in counts, from 1 - CPU_MAX_SPEED, and
* reported to the user in tenths of a percent.
*/
static u_int32_t cpu_duty_offset;
static u_int32_t cpu_duty_width;
#define CPU_MAX_SPEED (1 << cpu_duty_width)
#define CPU_SPEED_PERCENT(x) ((1000 * (x)) / CPU_MAX_SPEED)
#define CPU_SPEED_PRINTABLE(x) (CPU_SPEED_PERCENT(x) / 10),(CPU_SPEED_PERCENT(x) % 10)
static u_int32_t cpu_smi_cmd; /* should be a generic way to do this */
static u_int8_t cpu_pstate_cnt;
static u_int32_t cpu_current_state;
static u_int32_t cpu_performance_state;
static u_int32_t cpu_economy_state;
static u_int32_t cpu_max_state;
static device_t *cpu_devices;
static int cpu_ndevices;
static struct sysctl_ctx_list acpi_cpu_sysctl_ctx;
static struct sysctl_oid *acpi_cpu_sysctl_tree;
static int acpi_cpu_probe(device_t dev);
static int acpi_cpu_attach(device_t dev);
static void acpi_cpu_init_throttling(void *arg);
static void acpi_cpu_set_speed(u_int32_t speed);
static void acpi_cpu_power_profile(void *arg);
static int acpi_cpu_speed_sysctl(SYSCTL_HANDLER_ARGS);
static device_method_t acpi_cpu_methods[] = {
/* Device interface */
DEVMETHOD(device_probe, acpi_cpu_probe),
DEVMETHOD(device_attach, acpi_cpu_attach),
{0, 0}
};
static driver_t acpi_cpu_driver = {
"acpi_cpu",
acpi_cpu_methods,
sizeof(struct acpi_cpu_softc),
};
static devclass_t acpi_cpu_devclass;
DRIVER_MODULE(acpi_cpu, acpi, acpi_cpu_driver, acpi_cpu_devclass, 0, 0);
static int
acpi_cpu_probe(device_t dev)
{
if (!acpi_disabled("cpu") &&
(acpi_get_type(dev) == ACPI_TYPE_PROCESSOR)) {
device_set_desc(dev, "CPU"); /* XXX get more verbose description? */
return(0);
}
return(ENXIO);
}
static int
acpi_cpu_attach(device_t dev)
{
struct acpi_cpu_softc *sc;
struct acpi_softc *acpi_sc;
ACPI_OBJECT processor;
ACPI_BUFFER buf;
ACPI_STATUS status;
u_int32_t p_blk;
u_int32_t p_blk_length;
u_int32_t duty_end;
int rid;
ACPI_FUNCTION_TRACE((char *)(uintptr_t)__func__);
ACPI_ASSERTLOCK;
sc = device_get_softc(dev);
sc->cpu_dev = dev;
sc->cpu_handle = acpi_get_handle(dev);
/*
* Get global parameters from the FADT.
*/
if (device_get_unit(sc->cpu_dev) == 0) {
cpu_duty_offset = AcpiGbl_FADT->DutyOffset;
cpu_duty_width = AcpiGbl_FADT->DutyWidth;
cpu_smi_cmd = AcpiGbl_FADT->SmiCmd;
cpu_pstate_cnt = AcpiGbl_FADT->PstateCnt;
/* validate the offset/width */
if (cpu_duty_width > 0) {
duty_end = cpu_duty_offset + cpu_duty_width - 1;
/* check that it fits */
if (duty_end > 31) {
printf("acpi_cpu: CLK_VAL field overflows P_CNT register\n");
cpu_duty_width = 0;
}
/* check for overlap with the THT_EN bit */
if ((cpu_duty_offset <= 4) && (duty_end >= 4)) {
printf("acpi_cpu: CLK_VAL field overlaps THT_EN bit\n");
cpu_duty_width = 0;
}
}
/*
* Start the throttling process once the probe phase completes, if we think that
* it's going to be useful. If the duty width value is zero, there are no significant
* bits in the register and thus no throttled states.
*/
if (cpu_duty_width > 0) {
AcpiOsQueueForExecution(OSD_PRIORITY_LO, acpi_cpu_init_throttling, NULL);
acpi_sc = acpi_device_get_parent_softc(dev);
sysctl_ctx_init(&acpi_cpu_sysctl_ctx);
acpi_cpu_sysctl_tree = SYSCTL_ADD_NODE(&acpi_cpu_sysctl_ctx,
SYSCTL_CHILDREN(acpi_sc->acpi_sysctl_tree),
OID_AUTO, "cpu", CTLFLAG_RD, 0, "");
SYSCTL_ADD_INT(&acpi_cpu_sysctl_ctx, SYSCTL_CHILDREN(acpi_cpu_sysctl_tree),
OID_AUTO, "max_speed", CTLFLAG_RD,
&cpu_max_state, 0, "maximum CPU speed");
SYSCTL_ADD_INT(&acpi_cpu_sysctl_ctx, SYSCTL_CHILDREN(acpi_cpu_sysctl_tree),
OID_AUTO, "current_speed", CTLFLAG_RD,
&cpu_current_state, 0, "current CPU speed");
SYSCTL_ADD_PROC(&acpi_cpu_sysctl_ctx, SYSCTL_CHILDREN(acpi_cpu_sysctl_tree),
OID_AUTO, "performance_speed", CTLTYPE_INT | CTLFLAG_RW,
&cpu_performance_state, 0, acpi_cpu_speed_sysctl, "I", "");
SYSCTL_ADD_PROC(&acpi_cpu_sysctl_ctx, SYSCTL_CHILDREN(acpi_cpu_sysctl_tree),
OID_AUTO, "economy_speed", CTLTYPE_INT | CTLFLAG_RW,
&cpu_economy_state, 0, acpi_cpu_speed_sysctl, "I", "");
}
}
/*
* Get the processor object.
*/
buf.Pointer = &processor;
buf.Length = sizeof(processor);
if (ACPI_FAILURE(status = AcpiEvaluateObject(sc->cpu_handle, NULL, NULL, &buf))) {
device_printf(sc->cpu_dev, "couldn't get Processor object - %s\n", AcpiFormatException(status));
return_VALUE(ENXIO);
}
if (processor.Type != ACPI_TYPE_PROCESSOR) {
device_printf(sc->cpu_dev, "Processor object has bad type %d\n", processor.Type);
return_VALUE(ENXIO);
}
sc->cpu_id = processor.Processor.ProcId;
/*
* If it looks like we support throttling, find this CPU's P_BLK.
*
* Note that some systems seem to duplicate the P_BLK pointer across
* multiple CPUs, so not getting the resource is not fatal.
*
* XXX should support _PTC here as well, once we work out how to parse it.
*
* XXX is it valid to assume that the P_BLK must be 6 bytes long?
*/
if (cpu_duty_width > 0) {
p_blk = processor.Processor.PblkAddress;
p_blk_length = processor.Processor.PblkLength;
/* allocate bus space if possible */
if ((p_blk > 0) && (p_blk_length == 6)) {
rid = 0;
bus_set_resource(sc->cpu_dev, SYS_RES_IOPORT, rid, p_blk, p_blk_length);
sc->cpu_p_blk = bus_alloc_resource(sc->cpu_dev, SYS_RES_IOPORT, &rid, 0, ~0, 1,
RF_ACTIVE);
ACPI_DEBUG_PRINT((ACPI_DB_IO, "acpi_cpu%d: throttling with P_BLK at 0x%x/%d%s\n",
device_get_unit(sc->cpu_dev), p_blk, p_blk_length,
sc->cpu_p_blk ? "" : " (shadowed)"));
}
}
return_VALUE(0);
}
/*
* Call this *after* all CPUs have been attached.
*
* Takes the ACPI lock to avoid fighting anyone over the SMI command
* port. Could probably lock less code.
*/
static void
acpi_cpu_init_throttling(void *arg)
{
int cpu_temp_speed;
ACPI_LOCK_DECL;
ACPI_LOCK;
/* get set of CPU devices */
devclass_get_devices(acpi_cpu_devclass, &cpu_devices, &cpu_ndevices);
/* initialise throttling states */
cpu_max_state = CPU_MAX_SPEED;
cpu_performance_state = cpu_max_state;
cpu_economy_state = cpu_performance_state / 2;
if (cpu_economy_state == 0) /* 0 is 'reserved' */
cpu_economy_state++;
if (TUNABLE_INT_FETCH("hw.acpi.cpu.performance_speed",
&cpu_temp_speed) && cpu_temp_speed > 0 &&
cpu_temp_speed <= cpu_max_state)
cpu_performance_state = cpu_temp_speed;
if (TUNABLE_INT_FETCH("hw.acpi.cpu.economy_speed",
&cpu_temp_speed) && cpu_temp_speed > 0 &&
cpu_temp_speed <= cpu_max_state)
cpu_economy_state = cpu_temp_speed;
/* register performance profile change handler */
EVENTHANDLER_REGISTER(power_profile_change, acpi_cpu_power_profile, NULL, 0);
/* if ACPI 2.0+, signal platform that we are taking over throttling */
if (cpu_pstate_cnt != 0) {
/* XXX should be a generic interface for this */
AcpiOsWritePort(cpu_smi_cmd, cpu_pstate_cnt, 8);
}
ACPI_UNLOCK;
/* set initial speed */
acpi_cpu_power_profile(NULL);
printf("acpi_cpu: CPU throttling enabled, %d steps from 100%% to %d.%d%%\n",
CPU_MAX_SPEED, CPU_SPEED_PRINTABLE(1));
}
/*
* Set CPUs to the new state.
*
* Must be called with the ACPI lock held.
*/
static void
acpi_cpu_set_speed(u_int32_t speed)
{
struct acpi_cpu_softc *sc;
int i;
u_int32_t p_cnt, clk_val;
ACPI_ASSERTLOCK;
/* iterate over processors */
for (i = 0; i < cpu_ndevices; i++) {
sc = device_get_softc(cpu_devices[i]);
if (sc->cpu_p_blk == NULL)
continue;
/* get the current P_CNT value and disable throttling */
p_cnt = CPU_GET_P_CNT(sc);
p_cnt &= ~CPU_P_CNT_THT_EN;
CPU_SET_P_CNT(sc, p_cnt);
/* if we're at maximum speed, that's all */
if (speed < CPU_MAX_SPEED) {
/* mask the old CLK_VAL off and or-in the new value */
clk_val = CPU_MAX_SPEED << cpu_duty_offset;
p_cnt &= ~clk_val;
p_cnt |= (speed << cpu_duty_offset);
/* write the new P_CNT value and then enable throttling */
CPU_SET_P_CNT(sc, p_cnt);
p_cnt |= CPU_P_CNT_THT_EN;
CPU_SET_P_CNT(sc, p_cnt);
}
ACPI_VPRINT(sc->cpu_dev, acpi_device_get_parent_softc(sc->cpu_dev),
"set speed to %d.%d%%\n", CPU_SPEED_PRINTABLE(speed));
}
cpu_current_state = speed;
}
/*
* Power profile change hook.
*
* Uses the ACPI lock to avoid reentrancy.
*/
static void
acpi_cpu_power_profile(void *arg)
{
int state;
u_int32_t new;
ACPI_LOCK_DECL;
state = power_profile_get_state();
if (state != POWER_PROFILE_PERFORMANCE &&
state != POWER_PROFILE_ECONOMY) {
return;
}
ACPI_LOCK;
switch (state) {
case POWER_PROFILE_PERFORMANCE:
new = cpu_performance_state;
break;
case POWER_PROFILE_ECONOMY:
new = cpu_economy_state;
break;
default:
new = cpu_current_state;
break;
}
if (cpu_current_state != new)
acpi_cpu_set_speed(new);
ACPI_UNLOCK;
}
/*
* Handle changes in the performance/ecomony CPU settings.
*
* Does not need the ACPI lock (although setting *argp should
* probably be atomic).
*/
static int
acpi_cpu_speed_sysctl(SYSCTL_HANDLER_ARGS)
{
u_int32_t *argp;
u_int32_t arg;
int error;
argp = (u_int32_t *)oidp->oid_arg1;
arg = *argp;
error = sysctl_handle_int(oidp, &arg, 0, req);
/* error or no new value */
if ((error != 0) || (req->newptr == NULL))
return(error);
/* range check */
if ((arg < 1) || (arg > cpu_max_state))
return(EINVAL);
/* set new value and possibly switch */
*argp = arg;
acpi_cpu_power_profile(NULL);
return(0);
}