linux-stable/drivers/idle/intel_idle.c

1792 lines
46 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* intel_idle.c - native hardware idle loop for modern Intel processors
*
* Copyright (c) 2013 - 2020, Intel Corporation.
* Len Brown <len.brown@intel.com>
* Rafael J. Wysocki <rafael.j.wysocki@intel.com>
*/
/*
* intel_idle is a cpuidle driver that loads on all Intel CPUs with MWAIT
* in lieu of the legacy ACPI processor_idle driver. The intent is to
* make Linux more efficient on these processors, as intel_idle knows
* more than ACPI, as well as make Linux more immune to ACPI BIOS bugs.
*/
/*
* Design Assumptions
*
* All CPUs have same idle states as boot CPU
*
* Chipset BM_STS (bus master status) bit is a NOP
* for preventing entry into deep C-states
*
* CPU will flush caches as needed when entering a C-state via MWAIT
* (in contrast to entering ACPI C3, in which case the WBINVD
* instruction needs to be executed to flush the caches)
*/
/*
* Known limitations
*
* ACPI has a .suspend hack to turn off deep c-statees during suspend
* to avoid complications with the lapic timer workaround.
* Have not seen issues with suspend, but may need same workaround here.
*
*/
/* un-comment DEBUG to enable pr_debug() statements */
/* #define DEBUG */
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/acpi.h>
#include <linux/kernel.h>
#include <linux/cpuidle.h>
#include <linux/tick.h>
#include <trace/events/power.h>
#include <linux/sched.h>
#include <linux/notifier.h>
#include <linux/cpu.h>
drivers/idle: make intel_idle.c driver more explicitly non-modular The Kconfig for this driver is currently declared with: config INTEL_IDLE bool "Cpuidle Driver for Intel Processors" ...meaning that it currently is not being built as a module by anyone. This was done in commit 6ce9cd8669fa1195fdc21643370e34523c7ac988 ("intel_idle: disable module support") since "...the module capability is cauing more trouble than it is worth." This was done over 5y ago, and Daniel adds that: ...the modular support has been removed from almost all the cpuidle drivers and the cpuidle framework is no longer assuming driver could be unloaded. Removing the modular dead code in the driver makes sense as this what have been done in the others drivers. So lets remove the modular code that is essentially orphaned, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. At a later date we might want to consider whether subsys_init or another init category seems more appropriate than device_init. We replace module.h with moduleparam.h since the file does declare some module parameters, and leaving them as such is currently the easiest way to remain compatible with existing boot arg use cases. Note that MODULE_DEVICE_TABLE is a no-op for non-modular code. Also note that we can't remove intel_idle_cpuidle_devices_uninit() as that is still used for unwind purposes if the init fails. We also delete the MODULE_LICENSE tag etc. since all that information is already contained at the top of the file in the comments. Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Len Brown <len.brown@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-06-17 05:28:33 +00:00
#include <linux/moduleparam.h>
#include <asm/cpu_device_id.h>
#include <asm/intel-family.h>
#include <asm/mwait.h>
#include <asm/msr.h>
#define INTEL_IDLE_VERSION "0.5.1"
static struct cpuidle_driver intel_idle_driver = {
.name = "intel_idle",
.owner = THIS_MODULE,
};
/* intel_idle.max_cstate=0 disables driver */
static int max_cstate = CPUIDLE_STATE_MAX - 1;
static unsigned int disabled_states_mask;
static struct cpuidle_device __percpu *intel_idle_cpuidle_devices;
static unsigned long auto_demotion_disable_flags;
static bool disable_promotion_to_c1e;
struct idle_cpu {
struct cpuidle_state *state_table;
/*
* Hardware C-state auto-demotion may not always be optimal.
* Indicate which enable bits to clear here.
*/
unsigned long auto_demotion_disable_flags;
bool byt_auto_demotion_disable_flag;
bool disable_promotion_to_c1e;
bool use_acpi;
};
static const struct idle_cpu *icpu __initdata;
static struct cpuidle_state *cpuidle_state_table __initdata;
static unsigned int mwait_substates __initdata;
/*
* Enable this state by default even if the ACPI _CST does not list it.
*/
#define CPUIDLE_FLAG_ALWAYS_ENABLE BIT(15)
/*
* MWAIT takes an 8-bit "hint" in EAX "suggesting"
* the C-state (top nibble) and sub-state (bottom nibble)
* 0x00 means "MWAIT(C1)", 0x10 means "MWAIT(C2)" etc.
*
* We store the hint at the top of our "flags" for each state.
*/
#define flg2MWAIT(flags) (((flags) >> 24) & 0xFF)
#define MWAIT2flg(eax) ((eax & 0xFF) << 24)
/**
* intel_idle - Ask the processor to enter the given idle state.
* @dev: cpuidle device of the target CPU.
* @drv: cpuidle driver (assumed to point to intel_idle_driver).
* @index: Target idle state index.
*
* Use the MWAIT instruction to notify the processor that the CPU represented by
* @dev is idle and it can try to enter the idle state corresponding to @index.
*
* If the local APIC timer is not known to be reliable in the target idle state,
* enable one-shot tick broadcasting for the target CPU before executing MWAIT.
*
* Optionally call leave_mm() for the target CPU upfront to avoid wakeups due to
* flushing user TLBs.
*
* Must be called under local_irq_disable().
*/
static __cpuidle int intel_idle(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index)
{
struct cpuidle_state *state = &drv->states[index];
unsigned long eax = flg2MWAIT(state->flags);
unsigned long ecx = 1; /* break on interrupt flag */
mwait_idle_with_hints(eax, ecx);
return index;
}
/**
* intel_idle_s2idle - Ask the processor to enter the given idle state.
* @dev: cpuidle device of the target CPU.
* @drv: cpuidle driver (assumed to point to intel_idle_driver).
* @index: Target idle state index.
*
* Use the MWAIT instruction to notify the processor that the CPU represented by
* @dev is idle and it can try to enter the idle state corresponding to @index.
*
* Invoked as a suspend-to-idle callback routine with frozen user space, frozen
* scheduler tick and suspended scheduler clock on the target CPU.
*/
static __cpuidle int intel_idle_s2idle(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index)
{
unsigned long eax = flg2MWAIT(drv->states[index].flags);
unsigned long ecx = 1; /* break on interrupt flag */
mwait_idle_with_hints(eax, ecx);
return 0;
}
/*
* States are indexed by the cstate number,
* which is also the index into the MWAIT hint array.
* Thus C0 is a dummy.
*/
static struct cpuidle_state nehalem_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 3,
.target_residency = 6,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 20,
.target_residency = 80,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 200,
.target_residency = 800,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state snb_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 80,
.target_residency = 211,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 104,
.target_residency = 345,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7",
.desc = "MWAIT 0x30",
.flags = MWAIT2flg(0x30) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 109,
.target_residency = 345,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state byt_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6N",
.desc = "MWAIT 0x58",
.flags = MWAIT2flg(0x58) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 300,
.target_residency = 275,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6S",
.desc = "MWAIT 0x52",
.flags = MWAIT2flg(0x52) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 500,
.target_residency = 560,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 1200,
.target_residency = 4000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7S",
.desc = "MWAIT 0x64",
.flags = MWAIT2flg(0x64) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 10000,
.target_residency = 20000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state cht_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6N",
.desc = "MWAIT 0x58",
.flags = MWAIT2flg(0x58) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 80,
.target_residency = 275,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6S",
.desc = "MWAIT 0x52",
.flags = MWAIT2flg(0x52) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 200,
.target_residency = 560,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 1200,
.target_residency = 4000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7S",
.desc = "MWAIT 0x64",
.flags = MWAIT2flg(0x64) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 10000,
.target_residency = 20000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state ivb_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 59,
.target_residency = 156,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 80,
.target_residency = 300,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7",
.desc = "MWAIT 0x30",
.flags = MWAIT2flg(0x30) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 87,
.target_residency = 300,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state ivt_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 80,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 59,
.target_residency = 156,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 82,
.target_residency = 300,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state ivt_cstates_4s[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 250,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 59,
.target_residency = 300,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 84,
.target_residency = 400,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state ivt_cstates_8s[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 500,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 59,
.target_residency = 600,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 88,
.target_residency = 700,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state hsw_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 33,
.target_residency = 100,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 133,
.target_residency = 400,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7s",
.desc = "MWAIT 0x32",
.flags = MWAIT2flg(0x32) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 166,
.target_residency = 500,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C8",
.desc = "MWAIT 0x40",
.flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 300,
.target_residency = 900,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C9",
.desc = "MWAIT 0x50",
.flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 600,
.target_residency = 1800,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C10",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 2600,
.target_residency = 7700,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state bdw_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 40,
.target_residency = 100,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 133,
.target_residency = 400,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7s",
.desc = "MWAIT 0x32",
.flags = MWAIT2flg(0x32) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 166,
.target_residency = 500,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C8",
.desc = "MWAIT 0x40",
.flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 300,
.target_residency = 900,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C9",
.desc = "MWAIT 0x50",
.flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 600,
.target_residency = 1800,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C10",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 2600,
.target_residency = 7700,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state skl_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C3",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 70,
.target_residency = 100,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 85,
.target_residency = 200,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7s",
.desc = "MWAIT 0x33",
.flags = MWAIT2flg(0x33) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 124,
.target_residency = 800,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C8",
.desc = "MWAIT 0x40",
.flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 200,
.target_residency = 800,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C9",
.desc = "MWAIT 0x50",
.flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 480,
.target_residency = 5000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C10",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 890,
.target_residency = 5000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state skx_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 133,
.target_residency = 600,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state icx_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 1,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 4,
.target_residency = 4,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 170,
.target_residency = 600,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state atom_cstates[] __initdata = {
{
.name = "C1E",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C2",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10),
.exit_latency = 20,
.target_residency = 80,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C4",
.desc = "MWAIT 0x30",
.flags = MWAIT2flg(0x30) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 100,
.target_residency = 400,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x52",
.flags = MWAIT2flg(0x52) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 140,
.target_residency = 560,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state tangier_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 4,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C4",
.desc = "MWAIT 0x30",
.flags = MWAIT2flg(0x30) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 100,
.target_residency = 400,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x52",
.flags = MWAIT2flg(0x52) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 140,
.target_residency = 560,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 1200,
.target_residency = 4000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C9",
.desc = "MWAIT 0x64",
.flags = MWAIT2flg(0x64) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 10000,
.target_residency = 20000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state avn_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x51",
.flags = MWAIT2flg(0x51) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 15,
.target_residency = 45,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state knl_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 1,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle },
{
.name = "C6",
.desc = "MWAIT 0x10",
.flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 120,
.target_residency = 500,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle },
{
.enter = NULL }
};
static struct cpuidle_state bxt_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 133,
.target_residency = 133,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C7s",
.desc = "MWAIT 0x31",
.flags = MWAIT2flg(0x31) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 155,
.target_residency = 155,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C8",
.desc = "MWAIT 0x40",
.flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 1000,
.target_residency = 1000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C9",
.desc = "MWAIT 0x50",
.flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 2000,
.target_residency = 2000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C10",
.desc = "MWAIT 0x60",
.flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 10000,
.target_residency = 10000,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static struct cpuidle_state dnv_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 10,
.target_residency = 20,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 50,
.target_residency = 500,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
/*
* Note, depending on HW and FW revision, SnowRidge SoC may or may not support
* C6, and this is indicated in the CPUID mwait leaf.
*/
static struct cpuidle_state snr_cstates[] __initdata = {
{
.name = "C1",
.desc = "MWAIT 0x00",
.flags = MWAIT2flg(0x00),
.exit_latency = 2,
.target_residency = 2,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C1E",
.desc = "MWAIT 0x01",
.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
.exit_latency = 15,
.target_residency = 25,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.name = "C6",
.desc = "MWAIT 0x20",
.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 130,
.target_residency = 500,
.enter = &intel_idle,
.enter_s2idle = intel_idle_s2idle, },
{
.enter = NULL }
};
static const struct idle_cpu idle_cpu_nehalem __initconst = {
.state_table = nehalem_cstates,
.auto_demotion_disable_flags = NHM_C1_AUTO_DEMOTE | NHM_C3_AUTO_DEMOTE,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_nhx __initconst = {
.state_table = nehalem_cstates,
.auto_demotion_disable_flags = NHM_C1_AUTO_DEMOTE | NHM_C3_AUTO_DEMOTE,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_atom __initconst = {
.state_table = atom_cstates,
};
static const struct idle_cpu idle_cpu_tangier __initconst = {
.state_table = tangier_cstates,
};
static const struct idle_cpu idle_cpu_lincroft __initconst = {
.state_table = atom_cstates,
.auto_demotion_disable_flags = ATM_LNC_C6_AUTO_DEMOTE,
};
static const struct idle_cpu idle_cpu_snb __initconst = {
.state_table = snb_cstates,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_snx __initconst = {
.state_table = snb_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_byt __initconst = {
.state_table = byt_cstates,
.disable_promotion_to_c1e = true,
.byt_auto_demotion_disable_flag = true,
};
static const struct idle_cpu idle_cpu_cht __initconst = {
.state_table = cht_cstates,
.disable_promotion_to_c1e = true,
.byt_auto_demotion_disable_flag = true,
};
static const struct idle_cpu idle_cpu_ivb __initconst = {
.state_table = ivb_cstates,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_ivt __initconst = {
.state_table = ivt_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_hsw __initconst = {
.state_table = hsw_cstates,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_hsx __initconst = {
.state_table = hsw_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_bdw __initconst = {
.state_table = bdw_cstates,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_bdx __initconst = {
.state_table = bdw_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_skl __initconst = {
.state_table = skl_cstates,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_skx __initconst = {
.state_table = skx_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_icx __initconst = {
.state_table = icx_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_avn __initconst = {
.state_table = avn_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_knl __initconst = {
.state_table = knl_cstates,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_bxt __initconst = {
.state_table = bxt_cstates,
.disable_promotion_to_c1e = true,
};
static const struct idle_cpu idle_cpu_dnv __initconst = {
.state_table = dnv_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct idle_cpu idle_cpu_snr __initconst = {
.state_table = snr_cstates,
.disable_promotion_to_c1e = true,
.use_acpi = true,
};
static const struct x86_cpu_id intel_idle_ids[] __initconst = {
X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EP, &idle_cpu_nhx),
X86_MATCH_INTEL_FAM6_MODEL(NEHALEM, &idle_cpu_nehalem),
X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_G, &idle_cpu_nehalem),
X86_MATCH_INTEL_FAM6_MODEL(WESTMERE, &idle_cpu_nehalem),
X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EP, &idle_cpu_nhx),
X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EX, &idle_cpu_nhx),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_BONNELL, &idle_cpu_atom),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_BONNELL_MID, &idle_cpu_lincroft),
X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EX, &idle_cpu_nhx),
X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE, &idle_cpu_snb),
X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE_X, &idle_cpu_snx),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_SALTWELL, &idle_cpu_atom),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT, &idle_cpu_byt),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_MID, &idle_cpu_tangier),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_AIRMONT, &idle_cpu_cht),
X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE, &idle_cpu_ivb),
X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE_X, &idle_cpu_ivt),
X86_MATCH_INTEL_FAM6_MODEL(HASWELL, &idle_cpu_hsw),
X86_MATCH_INTEL_FAM6_MODEL(HASWELL_X, &idle_cpu_hsx),
X86_MATCH_INTEL_FAM6_MODEL(HASWELL_L, &idle_cpu_hsw),
X86_MATCH_INTEL_FAM6_MODEL(HASWELL_G, &idle_cpu_hsw),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D, &idle_cpu_avn),
X86_MATCH_INTEL_FAM6_MODEL(BROADWELL, &idle_cpu_bdw),
X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_G, &idle_cpu_bdw),
X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_X, &idle_cpu_bdx),
X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_D, &idle_cpu_bdx),
X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L, &idle_cpu_skl),
X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE, &idle_cpu_skl),
X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, &idle_cpu_skl),
X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, &idle_cpu_skl),
X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X, &idle_cpu_skx),
X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X, &idle_cpu_icx),
X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_D, &idle_cpu_icx),
X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL, &idle_cpu_knl),
X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM, &idle_cpu_knl),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT, &idle_cpu_bxt),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_PLUS, &idle_cpu_bxt),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D, &idle_cpu_dnv),
X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_D, &idle_cpu_snr),
{}
};
static const struct x86_cpu_id intel_mwait_ids[] __initconst = {
X86_MATCH_VENDOR_FAM_FEATURE(INTEL, 6, X86_FEATURE_MWAIT, NULL),
{}
};
static bool __init intel_idle_max_cstate_reached(int cstate)
{
if (cstate + 1 > max_cstate) {
pr_info("max_cstate %d reached\n", max_cstate);
return true;
}
return false;
}
static bool __init intel_idle_state_needs_timer_stop(struct cpuidle_state *state)
{
unsigned long eax = flg2MWAIT(state->flags);
if (boot_cpu_has(X86_FEATURE_ARAT))
return false;
/*
* Switch over to one-shot tick broadcast if the target C-state
* is deeper than C1.
*/
return !!((eax >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK);
}
#ifdef CONFIG_ACPI_PROCESSOR_CSTATE
#include <acpi/processor.h>
static bool no_acpi __read_mostly;
module_param(no_acpi, bool, 0444);
MODULE_PARM_DESC(no_acpi, "Do not use ACPI _CST for building the idle states list");
static bool force_use_acpi __read_mostly; /* No effect if no_acpi is set. */
module_param_named(use_acpi, force_use_acpi, bool, 0444);
MODULE_PARM_DESC(use_acpi, "Use ACPI _CST for building the idle states list");
static struct acpi_processor_power acpi_state_table __initdata;
/**
* intel_idle_cst_usable - Check if the _CST information can be used.
*
* Check if all of the C-states listed by _CST in the max_cstate range are
* ACPI_CSTATE_FFH, which means that they should be entered via MWAIT.
*/
static bool __init intel_idle_cst_usable(void)
{
int cstate, limit;
limit = min_t(int, min_t(int, CPUIDLE_STATE_MAX, max_cstate + 1),
acpi_state_table.count);
for (cstate = 1; cstate < limit; cstate++) {
struct acpi_processor_cx *cx = &acpi_state_table.states[cstate];
if (cx->entry_method != ACPI_CSTATE_FFH)
return false;
}
return true;
}
static bool __init intel_idle_acpi_cst_extract(void)
{
unsigned int cpu;
if (no_acpi) {
pr_debug("Not allowed to use ACPI _CST\n");
return false;
}
for_each_possible_cpu(cpu) {
struct acpi_processor *pr = per_cpu(processors, cpu);
if (!pr)
continue;
if (acpi_processor_evaluate_cst(pr->handle, cpu, &acpi_state_table))
continue;
acpi_state_table.count++;
if (!intel_idle_cst_usable())
continue;
intel_idle: Ignore _CST if control cannot be taken from the platform e6d4f08a6776 ("intel_idle: Use ACPI _CST on server systems") avoids enabling c-states that have been disabled by the platform with the exception of C1E. Unfortunately, BIOS implementations are not always consistent in terms of how capabilities are advertised and control cannot always be handed over. If control cannot be handed over then intel_idle reports that "ACPI _CST not found or not usable" but does not clear acpi_state_table.count meaning the information is still partially used. This patch ignores ACPI information if CST control cannot be requested from the platform. This was only observed on a number of Haswell platforms that had identical CPUs but not identical BIOS versions. While this problem may be rare overall, 24 separate test cases bisected to this specific commit across 4 separate test machines and is worth addressing. If the situation occurs, the kernel behaves as it did before commit e6d4f08a6776 and uses any c-states that are discovered. The affected test cases were all ones that involved a small number of processes -- exec microbenchmark, pipe microbenchmark, git test suite, netperf, tbench with one client and system call microbenchmark. Each case benefits from being able to use turboboost which is prevented if the lower c-states are unavailable. This may mask real regressions specific to older hardware so it is worth addressing. C-state status before and after the patch 5.9.0-vanilla POLL latency:0 disabled:0 default:enabled 5.9.0-vanilla C1 latency:2 disabled:0 default:enabled 5.9.0-vanilla C1E latency:10 disabled:0 default:enabled 5.9.0-vanilla C3 latency:33 disabled:1 default:disabled 5.9.0-vanilla C6 latency:133 disabled:1 default:disabled 5.9.0-ignore-cst-v1r1 POLL latency:0 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C1 latency:2 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C1E latency:10 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C3 latency:33 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C6 latency:133 disabled:0 default:enabled Patch enables C3/C6. Netperf UDP_STREAM netperf-udp 5.5.0 5.9.0 vanilla ignore-cst-v1r1 Hmean send-64 193.41 ( 0.00%) 226.54 * 17.13%* Hmean send-128 392.16 ( 0.00%) 450.54 * 14.89%* Hmean send-256 769.94 ( 0.00%) 881.85 * 14.53%* Hmean send-1024 2994.21 ( 0.00%) 3468.95 * 15.85%* Hmean send-2048 5725.60 ( 0.00%) 6628.99 * 15.78%* Hmean send-3312 8468.36 ( 0.00%) 10288.02 * 21.49%* Hmean send-4096 10135.46 ( 0.00%) 12387.57 * 22.22%* Hmean send-8192 17142.07 ( 0.00%) 19748.11 * 15.20%* Hmean send-16384 28539.71 ( 0.00%) 30084.45 * 5.41%* Fixes: e6d4f08a6776 ("intel_idle: Use ACPI _CST on server systems") Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Cc: 5.6+ <stable@vger.kernel.org> # 5.6+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2020-10-16 15:28:32 +00:00
if (!acpi_processor_claim_cst_control())
break;
return true;
}
intel_idle: Ignore _CST if control cannot be taken from the platform e6d4f08a6776 ("intel_idle: Use ACPI _CST on server systems") avoids enabling c-states that have been disabled by the platform with the exception of C1E. Unfortunately, BIOS implementations are not always consistent in terms of how capabilities are advertised and control cannot always be handed over. If control cannot be handed over then intel_idle reports that "ACPI _CST not found or not usable" but does not clear acpi_state_table.count meaning the information is still partially used. This patch ignores ACPI information if CST control cannot be requested from the platform. This was only observed on a number of Haswell platforms that had identical CPUs but not identical BIOS versions. While this problem may be rare overall, 24 separate test cases bisected to this specific commit across 4 separate test machines and is worth addressing. If the situation occurs, the kernel behaves as it did before commit e6d4f08a6776 and uses any c-states that are discovered. The affected test cases were all ones that involved a small number of processes -- exec microbenchmark, pipe microbenchmark, git test suite, netperf, tbench with one client and system call microbenchmark. Each case benefits from being able to use turboboost which is prevented if the lower c-states are unavailable. This may mask real regressions specific to older hardware so it is worth addressing. C-state status before and after the patch 5.9.0-vanilla POLL latency:0 disabled:0 default:enabled 5.9.0-vanilla C1 latency:2 disabled:0 default:enabled 5.9.0-vanilla C1E latency:10 disabled:0 default:enabled 5.9.0-vanilla C3 latency:33 disabled:1 default:disabled 5.9.0-vanilla C6 latency:133 disabled:1 default:disabled 5.9.0-ignore-cst-v1r1 POLL latency:0 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C1 latency:2 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C1E latency:10 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C3 latency:33 disabled:0 default:enabled 5.9.0-ignore-cst-v1r1 C6 latency:133 disabled:0 default:enabled Patch enables C3/C6. Netperf UDP_STREAM netperf-udp 5.5.0 5.9.0 vanilla ignore-cst-v1r1 Hmean send-64 193.41 ( 0.00%) 226.54 * 17.13%* Hmean send-128 392.16 ( 0.00%) 450.54 * 14.89%* Hmean send-256 769.94 ( 0.00%) 881.85 * 14.53%* Hmean send-1024 2994.21 ( 0.00%) 3468.95 * 15.85%* Hmean send-2048 5725.60 ( 0.00%) 6628.99 * 15.78%* Hmean send-3312 8468.36 ( 0.00%) 10288.02 * 21.49%* Hmean send-4096 10135.46 ( 0.00%) 12387.57 * 22.22%* Hmean send-8192 17142.07 ( 0.00%) 19748.11 * 15.20%* Hmean send-16384 28539.71 ( 0.00%) 30084.45 * 5.41%* Fixes: e6d4f08a6776 ("intel_idle: Use ACPI _CST on server systems") Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Cc: 5.6+ <stable@vger.kernel.org> # 5.6+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2020-10-16 15:28:32 +00:00
acpi_state_table.count = 0;
pr_debug("ACPI _CST not found or not usable\n");
return false;
}
static void __init intel_idle_init_cstates_acpi(struct cpuidle_driver *drv)
{
int cstate, limit = min_t(int, CPUIDLE_STATE_MAX, acpi_state_table.count);
/*
* If limit > 0, intel_idle_cst_usable() has returned 'true', so all of
* the interesting states are ACPI_CSTATE_FFH.
*/
for (cstate = 1; cstate < limit; cstate++) {
struct acpi_processor_cx *cx;
struct cpuidle_state *state;
if (intel_idle_max_cstate_reached(cstate - 1))
break;
cx = &acpi_state_table.states[cstate];
state = &drv->states[drv->state_count++];
snprintf(state->name, CPUIDLE_NAME_LEN, "C%d_ACPI", cstate);
strlcpy(state->desc, cx->desc, CPUIDLE_DESC_LEN);
state->exit_latency = cx->latency;
/*
* For C1-type C-states use the same number for both the exit
* latency and target residency, because that is the case for
* C1 in the majority of the static C-states tables above.
* For the other types of C-states, however, set the target
* residency to 3 times the exit latency which should lead to
* a reasonable balance between energy-efficiency and
* performance in the majority of interesting cases.
*/
state->target_residency = cx->latency;
if (cx->type > ACPI_STATE_C1)
state->target_residency *= 3;
state->flags = MWAIT2flg(cx->address);
if (cx->type > ACPI_STATE_C2)
state->flags |= CPUIDLE_FLAG_TLB_FLUSHED;
if (disabled_states_mask & BIT(cstate))
state->flags |= CPUIDLE_FLAG_OFF;
if (intel_idle_state_needs_timer_stop(state))
state->flags |= CPUIDLE_FLAG_TIMER_STOP;
state->enter = intel_idle;
state->enter_s2idle = intel_idle_s2idle;
}
}
static bool __init intel_idle_off_by_default(u32 mwait_hint)
{
int cstate, limit;
/*
* If there are no _CST C-states, do not disable any C-states by
* default.
*/
if (!acpi_state_table.count)
return false;
limit = min_t(int, CPUIDLE_STATE_MAX, acpi_state_table.count);
/*
* If limit > 0, intel_idle_cst_usable() has returned 'true', so all of
* the interesting states are ACPI_CSTATE_FFH.
*/
for (cstate = 1; cstate < limit; cstate++) {
if (acpi_state_table.states[cstate].address == mwait_hint)
return false;
}
return true;
}
#else /* !CONFIG_ACPI_PROCESSOR_CSTATE */
#define force_use_acpi (false)
static inline bool intel_idle_acpi_cst_extract(void) { return false; }
static inline void intel_idle_init_cstates_acpi(struct cpuidle_driver *drv) { }
static inline bool intel_idle_off_by_default(u32 mwait_hint) { return false; }
#endif /* !CONFIG_ACPI_PROCESSOR_CSTATE */
/**
* ivt_idle_state_table_update - Tune the idle states table for Ivy Town.
*
* Tune IVT multi-socket targets.
* Assumption: num_sockets == (max_package_num + 1).
*/
static void __init ivt_idle_state_table_update(void)
{
/* IVT uses a different table for 1-2, 3-4, and > 4 sockets */
int cpu, package_num, num_sockets = 1;
for_each_online_cpu(cpu) {
package_num = topology_physical_package_id(cpu);
if (package_num + 1 > num_sockets) {
num_sockets = package_num + 1;
if (num_sockets > 4) {
cpuidle_state_table = ivt_cstates_8s;
return;
}
}
}
if (num_sockets > 2)
cpuidle_state_table = ivt_cstates_4s;
/* else, 1 and 2 socket systems use default ivt_cstates */
}
/**
* irtl_2_usec - IRTL to microseconds conversion.
* @irtl: IRTL MSR value.
*
* Translate the IRTL (Interrupt Response Time Limit) MSR value to microseconds.
*/
static unsigned long long __init irtl_2_usec(unsigned long long irtl)
{
static const unsigned int irtl_ns_units[] __initconst = {
1, 32, 1024, 32768, 1048576, 33554432, 0, 0
};
unsigned long long ns;
if (!irtl)
return 0;
ns = irtl_ns_units[(irtl >> 10) & 0x7];
return div_u64((irtl & 0x3FF) * ns, NSEC_PER_USEC);
}
/**
* bxt_idle_state_table_update - Fix up the Broxton idle states table.
*
* On BXT, trust the IRTL (Interrupt Response Time Limit) MSR to show the
* definitive maximum latency and use the same value for target_residency.
*/
static void __init bxt_idle_state_table_update(void)
{
unsigned long long msr;
unsigned int usec;
rdmsrl(MSR_PKGC6_IRTL, msr);
usec = irtl_2_usec(msr);
if (usec) {
bxt_cstates[2].exit_latency = usec;
bxt_cstates[2].target_residency = usec;
}
rdmsrl(MSR_PKGC7_IRTL, msr);
usec = irtl_2_usec(msr);
if (usec) {
bxt_cstates[3].exit_latency = usec;
bxt_cstates[3].target_residency = usec;
}
rdmsrl(MSR_PKGC8_IRTL, msr);
usec = irtl_2_usec(msr);
if (usec) {
bxt_cstates[4].exit_latency = usec;
bxt_cstates[4].target_residency = usec;
}
rdmsrl(MSR_PKGC9_IRTL, msr);
usec = irtl_2_usec(msr);
if (usec) {
bxt_cstates[5].exit_latency = usec;
bxt_cstates[5].target_residency = usec;
}
rdmsrl(MSR_PKGC10_IRTL, msr);
usec = irtl_2_usec(msr);
if (usec) {
bxt_cstates[6].exit_latency = usec;
bxt_cstates[6].target_residency = usec;
}
}
/**
* sklh_idle_state_table_update - Fix up the Sky Lake idle states table.
*
* On SKL-H (model 0x5e) skip C8 and C9 if C10 is enabled and SGX disabled.
*/
static void __init sklh_idle_state_table_update(void)
{
unsigned long long msr;
unsigned int eax, ebx, ecx, edx;
/* if PC10 disabled via cmdline intel_idle.max_cstate=7 or shallower */
if (max_cstate <= 7)
return;
/* if PC10 not present in CPUID.MWAIT.EDX */
if ((mwait_substates & (0xF << 28)) == 0)
return;
rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr);
/* PC10 is not enabled in PKG C-state limit */
if ((msr & 0xF) != 8)
return;
ecx = 0;
cpuid(7, &eax, &ebx, &ecx, &edx);
/* if SGX is present */
if (ebx & (1 << 2)) {
x86/msr-index: Clean up bit defines for IA32_FEATURE_CONTROL MSR As pointed out by Boris, the defines for bits in IA32_FEATURE_CONTROL are quite a mouthful, especially the VMX bits which must differentiate between enabling VMX inside and outside SMX (TXT) operation. Rename the MSR and its bit defines to abbreviate FEATURE_CONTROL as FEAT_CTL to make them a little friendlier on the eyes. Arguably, the MSR itself should keep the full IA32_FEATURE_CONTROL name to match Intel's SDM, but a future patch will add a dedicated Kconfig, file and functions for the MSR. Using the full name for those assets is rather unwieldy, so bite the bullet and use IA32_FEAT_CTL so that its nomenclature is consistent throughout the kernel. Opportunistically, fix a few other annoyances with the defines: - Relocate the bit defines so that they immediately follow the MSR define, e.g. aren't mistaken as belonging to MISC_FEATURE_CONTROL. - Add whitespace around the block of feature control defines to make it clear they're all related. - Use BIT() instead of manually encoding the bit shift. - Use "VMX" instead of "VMXON" to match the SDM. - Append "_ENABLED" to the LMCE (Local Machine Check Exception) bit to be consistent with the kernel's verbiage used for all other feature control bits. Note, the SDM refers to the LMCE bit as LMCE_ON, likely to differentiate it from IA32_MCG_EXT_CTL.LMCE_EN. Ignore the (literal) one-off usage of _ON, the SDM is simply "wrong". Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Borislav Petkov <bp@suse.de> Link: https://lkml.kernel.org/r/20191221044513.21680-2-sean.j.christopherson@intel.com
2019-12-21 04:44:55 +00:00
rdmsrl(MSR_IA32_FEAT_CTL, msr);
/* if SGX is enabled */
if (msr & (1 << 18))
return;
}
skl_cstates[5].flags |= CPUIDLE_FLAG_UNUSABLE; /* C8-SKL */
skl_cstates[6].flags |= CPUIDLE_FLAG_UNUSABLE; /* C9-SKL */
}
intel_idle: Adjust the SKX C6 parameters if PC6 is disabled Because cpuidle assumes worst-case C-state parameters, PC6 parameters are used for describing C6, which is worst-case for requesting CC6. When PC6 is enabled, this is appropriate. But if PC6 is disabled in the BIOS, the exit latency and target residency should be adjusted accordingly. Exit latency: Previously the C6 exit latency was measured as the PC6 exit latency. With PC6 disabled, the C6 exit latency should be the one of CC6. Target residency: With PC6 disabled, the idle duration within [CC6, PC6) would make the idle governor choose C1E over C6. This would cause low energy-efficiency. We should lower the bar to request C6 when PC6 is disabled. To fill this gap, check if PC6 is disabled in the BIOS in the MSR_PKG_CST_CONFIG_CONTROL(0xe2) register. If so, use the CC6 exit latency for C6 and set target_residency to 3 times of the new exit latency. [This is consistent with how intel_idle driver uses _CST to calculate the target_residency.] As a result, the OS would be more likely to choose C6 over C1E when PC6 is disabled, which is reasonable, because if C6 is enabled, it implies that the user cares about energy, so choosing C6 more frequently makes sense. The new CC6 exit latency of 92us was measured with wult[1] on SKX via NIC wakeup as the 99.99th percentile. Also CLX and CPX both have the same CPU model number as SkX, but their CC6 exit latencies are similar to the SKX one, 96us and 89us respectively, so reuse the SKX value for them. There is a concern that it might be better to use a more generic approach instead of optimizing every platform. However, if the required code complexity and different PC6 bit interpretation on different platforms are taken into account, tuning the code per platform seems to be an acceptable tradeoff. Link: https://intel.github.io/wult/ # [1] Suggested-by: Len Brown <len.brown@intel.com> Signed-off-by: Chen Yu <yu.c.chen@intel.com> Reviewed-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com> [ rjw: Subject and changelog edits ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2021-05-28 03:20:54 +00:00
/**
* skx_idle_state_table_update - Adjust the Sky Lake/Cascade Lake
* idle states table.
*/
static void __init skx_idle_state_table_update(void)
{
unsigned long long msr;
rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr);
/*
* 000b: C0/C1 (no package C-state support)
* 001b: C2
* 010b: C6 (non-retention)
* 011b: C6 (retention)
* 111b: No Package C state limits.
*/
if ((msr & 0x7) < 2) {
/*
* Uses the CC6 + PC0 latency and 3 times of
* latency for target_residency if the PC6
* is disabled in BIOS. This is consistent
* with how intel_idle driver uses _CST
* to set the target_residency.
*/
skx_cstates[2].exit_latency = 92;
skx_cstates[2].target_residency = 276;
}
}
static bool __init intel_idle_verify_cstate(unsigned int mwait_hint)
{
unsigned int mwait_cstate = MWAIT_HINT2CSTATE(mwait_hint) + 1;
unsigned int num_substates = (mwait_substates >> mwait_cstate * 4) &
MWAIT_SUBSTATE_MASK;
/* Ignore the C-state if there are NO sub-states in CPUID for it. */
if (num_substates == 0)
return false;
if (mwait_cstate > 2 && !boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
mark_tsc_unstable("TSC halts in idle states deeper than C2");
return true;
}
static void __init intel_idle_init_cstates_icpu(struct cpuidle_driver *drv)
{
int cstate;
switch (boot_cpu_data.x86_model) {
case INTEL_FAM6_IVYBRIDGE_X:
ivt_idle_state_table_update();
break;
case INTEL_FAM6_ATOM_GOLDMONT:
case INTEL_FAM6_ATOM_GOLDMONT_PLUS:
bxt_idle_state_table_update();
break;
case INTEL_FAM6_SKYLAKE:
sklh_idle_state_table_update();
break;
intel_idle: Adjust the SKX C6 parameters if PC6 is disabled Because cpuidle assumes worst-case C-state parameters, PC6 parameters are used for describing C6, which is worst-case for requesting CC6. When PC6 is enabled, this is appropriate. But if PC6 is disabled in the BIOS, the exit latency and target residency should be adjusted accordingly. Exit latency: Previously the C6 exit latency was measured as the PC6 exit latency. With PC6 disabled, the C6 exit latency should be the one of CC6. Target residency: With PC6 disabled, the idle duration within [CC6, PC6) would make the idle governor choose C1E over C6. This would cause low energy-efficiency. We should lower the bar to request C6 when PC6 is disabled. To fill this gap, check if PC6 is disabled in the BIOS in the MSR_PKG_CST_CONFIG_CONTROL(0xe2) register. If so, use the CC6 exit latency for C6 and set target_residency to 3 times of the new exit latency. [This is consistent with how intel_idle driver uses _CST to calculate the target_residency.] As a result, the OS would be more likely to choose C6 over C1E when PC6 is disabled, which is reasonable, because if C6 is enabled, it implies that the user cares about energy, so choosing C6 more frequently makes sense. The new CC6 exit latency of 92us was measured with wult[1] on SKX via NIC wakeup as the 99.99th percentile. Also CLX and CPX both have the same CPU model number as SkX, but their CC6 exit latencies are similar to the SKX one, 96us and 89us respectively, so reuse the SKX value for them. There is a concern that it might be better to use a more generic approach instead of optimizing every platform. However, if the required code complexity and different PC6 bit interpretation on different platforms are taken into account, tuning the code per platform seems to be an acceptable tradeoff. Link: https://intel.github.io/wult/ # [1] Suggested-by: Len Brown <len.brown@intel.com> Signed-off-by: Chen Yu <yu.c.chen@intel.com> Reviewed-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com> [ rjw: Subject and changelog edits ] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2021-05-28 03:20:54 +00:00
case INTEL_FAM6_SKYLAKE_X:
skx_idle_state_table_update();
break;
}
for (cstate = 0; cstate < CPUIDLE_STATE_MAX; ++cstate) {
unsigned int mwait_hint;
if (intel_idle_max_cstate_reached(cstate))
break;
if (!cpuidle_state_table[cstate].enter &&
!cpuidle_state_table[cstate].enter_s2idle)
break;
/* If marked as unusable, skip this state. */
if (cpuidle_state_table[cstate].flags & CPUIDLE_FLAG_UNUSABLE) {
pr_debug("state %s is disabled\n",
cpuidle_state_table[cstate].name);
continue;
}
mwait_hint = flg2MWAIT(cpuidle_state_table[cstate].flags);
if (!intel_idle_verify_cstate(mwait_hint))
continue;
/* Structure copy. */
drv->states[drv->state_count] = cpuidle_state_table[cstate];
if ((disabled_states_mask & BIT(drv->state_count)) ||
((icpu->use_acpi || force_use_acpi) &&
intel_idle_off_by_default(mwait_hint) &&
!(cpuidle_state_table[cstate].flags & CPUIDLE_FLAG_ALWAYS_ENABLE)))
drv->states[drv->state_count].flags |= CPUIDLE_FLAG_OFF;
if (intel_idle_state_needs_timer_stop(&drv->states[drv->state_count]))
drv->states[drv->state_count].flags |= CPUIDLE_FLAG_TIMER_STOP;
drv->state_count++;
}
if (icpu->byt_auto_demotion_disable_flag) {
wrmsrl(MSR_CC6_DEMOTION_POLICY_CONFIG, 0);
wrmsrl(MSR_MC6_DEMOTION_POLICY_CONFIG, 0);
}
}
/**
* intel_idle_cpuidle_driver_init - Create the list of available idle states.
* @drv: cpuidle driver structure to initialize.
*/
static void __init intel_idle_cpuidle_driver_init(struct cpuidle_driver *drv)
{
cpuidle_poll_state_init(drv);
if (disabled_states_mask & BIT(0))
drv->states[0].flags |= CPUIDLE_FLAG_OFF;
drv->state_count = 1;
if (icpu)
intel_idle_init_cstates_icpu(drv);
else
intel_idle_init_cstates_acpi(drv);
}
static void auto_demotion_disable(void)
{
unsigned long long msr_bits;
rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
msr_bits &= ~auto_demotion_disable_flags;
wrmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
}
static void c1e_promotion_disable(void)
{
unsigned long long msr_bits;
rdmsrl(MSR_IA32_POWER_CTL, msr_bits);
msr_bits &= ~0x2;
wrmsrl(MSR_IA32_POWER_CTL, msr_bits);
}
/**
* intel_idle_cpu_init - Register the target CPU with the cpuidle core.
* @cpu: CPU to initialize.
*
* Register a cpuidle device object for @cpu and update its MSRs in accordance
* with the processor model flags.
*/
static int intel_idle_cpu_init(unsigned int cpu)
{
struct cpuidle_device *dev;
dev = per_cpu_ptr(intel_idle_cpuidle_devices, cpu);
dev->cpu = cpu;
if (cpuidle_register_device(dev)) {
pr_debug("cpuidle_register_device %d failed!\n", cpu);
return -EIO;
}
if (auto_demotion_disable_flags)
auto_demotion_disable();
if (disable_promotion_to_c1e)
c1e_promotion_disable();
return 0;
}
static int intel_idle_cpu_online(unsigned int cpu)
{
struct cpuidle_device *dev;
if (!boot_cpu_has(X86_FEATURE_ARAT))
tick_broadcast_enable();
/*
* Some systems can hotplug a cpu at runtime after
* the kernel has booted, we have to initialize the
* driver in this case
*/
dev = per_cpu_ptr(intel_idle_cpuidle_devices, cpu);
if (!dev->registered)
return intel_idle_cpu_init(cpu);
return 0;
}
/**
* intel_idle_cpuidle_devices_uninit - Unregister all cpuidle devices.
*/
static void __init intel_idle_cpuidle_devices_uninit(void)
{
int i;
for_each_online_cpu(i)
cpuidle_unregister_device(per_cpu_ptr(intel_idle_cpuidle_devices, i));
}
static int __init intel_idle_init(void)
{
const struct x86_cpu_id *id;
unsigned int eax, ebx, ecx;
int retval;
/* Do not load intel_idle at all for now if idle= is passed */
if (boot_option_idle_override != IDLE_NO_OVERRIDE)
return -ENODEV;
if (max_cstate == 0) {
pr_debug("disabled\n");
return -EPERM;
}
id = x86_match_cpu(intel_idle_ids);
if (id) {
if (!boot_cpu_has(X86_FEATURE_MWAIT)) {
pr_debug("Please enable MWAIT in BIOS SETUP\n");
return -ENODEV;
}
} else {
id = x86_match_cpu(intel_mwait_ids);
if (!id)
return -ENODEV;
}
if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
return -ENODEV;
cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &mwait_substates);
if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) ||
!(ecx & CPUID5_ECX_INTERRUPT_BREAK) ||
!mwait_substates)
return -ENODEV;
pr_debug("MWAIT substates: 0x%x\n", mwait_substates);
icpu = (const struct idle_cpu *)id->driver_data;
if (icpu) {
cpuidle_state_table = icpu->state_table;
auto_demotion_disable_flags = icpu->auto_demotion_disable_flags;
disable_promotion_to_c1e = icpu->disable_promotion_to_c1e;
if (icpu->use_acpi || force_use_acpi)
intel_idle_acpi_cst_extract();
} else if (!intel_idle_acpi_cst_extract()) {
return -ENODEV;
}
pr_debug("v" INTEL_IDLE_VERSION " model 0x%X\n",
boot_cpu_data.x86_model);
intel_idle_cpuidle_devices = alloc_percpu(struct cpuidle_device);
if (!intel_idle_cpuidle_devices)
return -ENOMEM;
intel_idle_cpuidle_driver_init(&intel_idle_driver);
retval = cpuidle_register_driver(&intel_idle_driver);
if (retval) {
struct cpuidle_driver *drv = cpuidle_get_driver();
printk(KERN_DEBUG pr_fmt("intel_idle yielding to %s\n"),
drv ? drv->name : "none");
goto init_driver_fail;
}
retval = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "idle/intel:online",
intel_idle_cpu_online, NULL);
if (retval < 0)
goto hp_setup_fail;
pr_debug("Local APIC timer is reliable in %s\n",
boot_cpu_has(X86_FEATURE_ARAT) ? "all C-states" : "C1");
return 0;
hp_setup_fail:
intel_idle_cpuidle_devices_uninit();
cpuidle_unregister_driver(&intel_idle_driver);
init_driver_fail:
free_percpu(intel_idle_cpuidle_devices);
return retval;
}
drivers/idle: make intel_idle.c driver more explicitly non-modular The Kconfig for this driver is currently declared with: config INTEL_IDLE bool "Cpuidle Driver for Intel Processors" ...meaning that it currently is not being built as a module by anyone. This was done in commit 6ce9cd8669fa1195fdc21643370e34523c7ac988 ("intel_idle: disable module support") since "...the module capability is cauing more trouble than it is worth." This was done over 5y ago, and Daniel adds that: ...the modular support has been removed from almost all the cpuidle drivers and the cpuidle framework is no longer assuming driver could be unloaded. Removing the modular dead code in the driver makes sense as this what have been done in the others drivers. So lets remove the modular code that is essentially orphaned, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. At a later date we might want to consider whether subsys_init or another init category seems more appropriate than device_init. We replace module.h with moduleparam.h since the file does declare some module parameters, and leaving them as such is currently the easiest way to remain compatible with existing boot arg use cases. Note that MODULE_DEVICE_TABLE is a no-op for non-modular code. Also note that we can't remove intel_idle_cpuidle_devices_uninit() as that is still used for unwind purposes if the init fails. We also delete the MODULE_LICENSE tag etc. since all that information is already contained at the top of the file in the comments. Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Len Brown <len.brown@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-06-17 05:28:33 +00:00
device_initcall(intel_idle_init);
drivers/idle: make intel_idle.c driver more explicitly non-modular The Kconfig for this driver is currently declared with: config INTEL_IDLE bool "Cpuidle Driver for Intel Processors" ...meaning that it currently is not being built as a module by anyone. This was done in commit 6ce9cd8669fa1195fdc21643370e34523c7ac988 ("intel_idle: disable module support") since "...the module capability is cauing more trouble than it is worth." This was done over 5y ago, and Daniel adds that: ...the modular support has been removed from almost all the cpuidle drivers and the cpuidle framework is no longer assuming driver could be unloaded. Removing the modular dead code in the driver makes sense as this what have been done in the others drivers. So lets remove the modular code that is essentially orphaned, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. At a later date we might want to consider whether subsys_init or another init category seems more appropriate than device_init. We replace module.h with moduleparam.h since the file does declare some module parameters, and leaving them as such is currently the easiest way to remain compatible with existing boot arg use cases. Note that MODULE_DEVICE_TABLE is a no-op for non-modular code. Also note that we can't remove intel_idle_cpuidle_devices_uninit() as that is still used for unwind purposes if the init fails. We also delete the MODULE_LICENSE tag etc. since all that information is already contained at the top of the file in the comments. Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Len Brown <len.brown@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-06-17 05:28:33 +00:00
/*
* We are not really modular, but we used to support that. Meaning we also
* support "intel_idle.max_cstate=..." at boot and also a read-only export of
* it at /sys/module/intel_idle/parameters/max_cstate -- so using module_param
* is the easiest way (currently) to continue doing that.
*/
module_param(max_cstate, int, 0444);
/*
* The positions of the bits that are set in this number are the indices of the
* idle states to be disabled by default (as reflected by the names of the
* corresponding idle state directories in sysfs, "state0", "state1" ...
* "state<i>" ..., where <i> is the index of the given state).
*/
module_param_named(states_off, disabled_states_mask, uint, 0444);
MODULE_PARM_DESC(states_off, "Mask of disabled idle states");