From: Len Brown <lenb@kernel.org>
To: linux-acpi@vger.kernel.org, linux-pm@lists.linux-foundation.org
Cc: linux-kernel@vger.kernel.org,
Boris Ostrovsky <boris.ostrovsky@amd.com>,
Len Brown <len.brown@intel.com>
Subject: [PATCH 58/76] idle, x86: Allow off-lined CPU to enter deeper C states
Date: Fri, 30 Mar 2012 06:14:01 -0400 [thread overview]
Message-ID: <1a022e3f1be11730bd8747b1af96a0274bf6356e.1333101989.git.len.brown@intel.com> (raw)
In-Reply-To: <1333102459-23750-1-git-send-email-lenb@kernel.org>
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Boris Ostrovsky <boris.ostrovsky@amd.com>
Currently when a CPU is off-lined it enters either MWAIT-based idle or,
if MWAIT is not desired or supported, HLT-based idle (which places the
processor in C1 state). This patch allows processors without MWAIT
support to stay in states deeper than C1.
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@amd.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
arch/x86/kernel/smpboot.c | 4 +++-
drivers/acpi/processor_idle.c | 31 +++++++++++++++++++++++++++++++
drivers/cpuidle/cpuidle.c | 28 ++++++++++++++++++++++++++++
include/linux/cpuidle.h | 5 +++++
4 files changed, 67 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 66d250c..93a2a09 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -50,6 +50,7 @@
#include <linux/tboot.h>
#include <linux/stackprotector.h>
#include <linux/gfp.h>
+#include <linux/cpuidle.h>
#include <asm/acpi.h>
#include <asm/desc.h>
@@ -1422,7 +1423,8 @@ void native_play_dead(void)
tboot_shutdown(TB_SHUTDOWN_WFS);
mwait_play_dead(); /* Only returns on failure */
- hlt_play_dead();
+ if (cpuidle_play_dead())
+ hlt_play_dead();
}
#else /* ... !CONFIG_HOTPLUG_CPU */
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 0e8e2de..6b1d32a 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -770,6 +770,35 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev,
return index;
}
+
+/**
+ * acpi_idle_play_dead - enters an ACPI state for long-term idle (i.e. off-lining)
+ * @dev: the target CPU
+ * @index: the index of suggested state
+ */
+static int acpi_idle_play_dead(struct cpuidle_device *dev, int index)
+{
+ struct cpuidle_state_usage *state_usage = &dev->states_usage[index];
+ struct acpi_processor_cx *cx = cpuidle_get_statedata(state_usage);
+
+ ACPI_FLUSH_CPU_CACHE();
+
+ while (1) {
+
+ if (cx->entry_method == ACPI_CSTATE_HALT)
+ halt();
+ else if (cx->entry_method == ACPI_CSTATE_SYSTEMIO) {
+ inb(cx->address);
+ /* See comment in acpi_idle_do_entry() */
+ inl(acpi_gbl_FADT.xpm_timer_block.address);
+ } else
+ return -ENODEV;
+ }
+
+ /* Never reached */
+ return 0;
+}
+
/**
* acpi_idle_enter_simple - enters an ACPI state without BM handling
* @dev: the target CPU
@@ -1077,12 +1106,14 @@ static int acpi_processor_setup_cpuidle_states(struct acpi_processor *pr)
state->flags |= CPUIDLE_FLAG_TIME_VALID;
state->enter = acpi_idle_enter_c1;
+ state->enter_dead = acpi_idle_play_dead;
drv->safe_state_index = count;
break;
case ACPI_STATE_C2:
state->flags |= CPUIDLE_FLAG_TIME_VALID;
state->enter = acpi_idle_enter_simple;
+ state->enter_dead = acpi_idle_play_dead;
drv->safe_state_index = count;
break;
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index f7cab5e..3e146b2 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -72,6 +72,34 @@ typedef int (*cpuidle_enter_t)(struct cpuidle_device *dev,
static cpuidle_enter_t cpuidle_enter_ops;
/**
+ * cpuidle_play_dead - cpu off-lining
+ *
+ * Only returns in case of an error
+ */
+int cpuidle_play_dead(void)
+{
+ struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices);
+ struct cpuidle_driver *drv = cpuidle_get_driver();
+ int i, dead_state = -1;
+ int power_usage = -1;
+
+ /* Find lowest-power state that supports long-term idle */
+ for (i = CPUIDLE_DRIVER_STATE_START; i < drv->state_count; i++) {
+ struct cpuidle_state *s = &drv->states[i];
+
+ if (s->power_usage < power_usage && s->enter_dead) {
+ power_usage = s->power_usage;
+ dead_state = i;
+ }
+ }
+
+ if (dead_state != -1)
+ return drv->states[dead_state].enter_dead(dev, dead_state);
+
+ return -ENODEV;
+}
+
+/**
* cpuidle_idle_call - the main idle loop
*
* NOTE: no locks or semaphores should be used here
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index f3ebbba..d557bcd 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -51,6 +51,8 @@ struct cpuidle_state {
int (*enter) (struct cpuidle_device *dev,
struct cpuidle_driver *drv,
int index);
+
+ int (*enter_dead) (struct cpuidle_device *dev, int index);
};
/* Idle State Flags */
@@ -147,6 +149,8 @@ extern int cpuidle_wrap_enter(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index,
int (*enter)(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index));
+extern int cpuidle_play_dead(void);
+
#else
static inline void disable_cpuidle(void) { }
static inline int cpuidle_idle_call(void) { return -ENODEV; }
@@ -168,6 +172,7 @@ static inline int cpuidle_wrap_enter(struct cpuidle_device *dev,
int (*enter)(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index))
{ return -ENODEV; }
+static inline int cpuidle_play_dead(void) {return -ENODEV; }
#endif
--
1.7.10.rc2.19.gfae9d
next prev parent reply other threads:[~2012-03-30 11:22 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-30 10:13 ACPI & Power Management patches for Linux-3.4 Len Brown
2012-03-30 10:13 ` [PATCH 01/76] x86, acpi, tboot: Have a ACPI os prepare sleep instead of calling tboot_sleep Len Brown
2012-03-30 10:13 ` [PATCH 02/76] tboot: Add return values for tboot_sleep Len Brown
2012-03-30 10:13 ` [PATCH 03/76] ACPI: ignore FADT reset-reg-sup flag Len Brown
2012-03-30 10:13 ` [PATCH 04/76] ACPICA: Fix regression in FADT revision checks Len Brown
2012-03-30 13:14 ` Josh Boyer
2012-04-03 19:58 ` [3.0.y, 3.2.y, 3.3.y] " Jonathan Nieder
2012-04-03 20:15 ` Josh Boyer
2012-04-04 18:58 ` Greg Kroah-Hartman
2012-03-30 10:13 ` [PATCH 05/76] cpuidle: Add common time keeping and irq enabling Len Brown
2012-03-30 10:13 ` [PATCH 06/76] ARM: at91: Consolidate time keeping and irq enable Len Brown
2012-03-30 10:13 ` [PATCH 07/76] ARM: kirkwood: " Len Brown
2012-03-30 10:13 ` [PATCH 08/76] ARM: davinci: " Len Brown
2012-03-30 10:13 ` [PATCH 09/76] ARM: omap: Consolidate OMAP3 " Len Brown
2012-03-30 10:13 ` [PATCH 10/76] ARM: omap: Consolidate OMAP4 " Len Brown
2012-03-30 10:13 ` [PATCH 11/76] ARM: shmobile: Consolidate " Len Brown
2012-03-30 10:13 ` [PATCH 12/76] SH: " Len Brown
2012-03-30 10:13 ` [PATCH 13/76] drivers/thermal/thermal_sys.c: fix build warning Len Brown
2012-03-30 10:13 ` [PATCH 14/76] thermal_sys: remove unnecessary line continuations Len Brown
2012-03-30 10:13 ` [PATCH 15/76] thermal_sys: remove obfuscating used-once macros Len Brown
2012-03-30 10:13 ` [PATCH 16/76] thermal_sys: kernel style cleanups Len Brown
2012-03-30 10:13 ` [PATCH 17/76] thermal_sys: convert printks to pr_<level> Len Brown
2012-03-30 13:41 ` [linux-pm] " Eduardo Valentin
2012-03-30 19:08 ` Joe Perches
2012-04-01 19:13 ` Eduardo Valentin
2012-03-30 10:13 ` [PATCH 18/76] thermal: add support for thermal sensor present on SPEAr13xx machines Len Brown
2012-03-30 10:13 ` [PATCH 19/76] thermal/spear_thermal: replace readl/writel with lighter _relaxed variants Len Brown
2012-03-30 10:13 ` [PATCH 20/76] thermal: spear13xx: checking for NULL instead of IS_ERR() Len Brown
2012-03-30 10:13 ` [PATCH 21/76] thermal: Fix for setting the thermal zone mode to enable/disable Len Brown
2012-03-30 10:13 ` [PATCH 22/76] ARM: davinci: Fix for cpuidle consolidation changes Len Brown
2012-03-30 10:13 ` [PATCH 23/76] ACPICA: Update _REV return value to 5 Len Brown
2012-03-30 10:13 ` [PATCH 24/76] ACPICA: ACPI 5: Support for new FADT SleepStatus, SleepControl registers Len Brown
2012-03-30 10:13 ` [PATCH 25/76] ACPICA: Move ACPI timer prototypes to public acpixf file Len Brown
2012-03-30 10:13 ` [PATCH 26/76] ACPICA: Support for custom ACPICA build for ACPI 5 reduced hardware Len Brown
2012-03-30 10:13 ` [PATCH 27/76] ACPICA: Expand OSL memory read/write interfaces to 64 bits Len Brown
2012-03-30 10:13 ` [PATCH 28/76] ACPICA: ACPI 5: Update debug output for new notify values Len Brown
2012-03-30 10:13 ` [PATCH 29/76] ACPICA: Add acpi_os_physical_table_override interface Len Brown
2012-03-30 10:13 ` [PATCH 30/76] ACPICA: Distill multiple sleep method functions to a single function Len Brown
2012-03-30 10:13 ` [PATCH 31/76] ACPICA: Split sleep/wake functions into two files Len Brown
2012-03-30 10:13 ` [PATCH 32/76] ACPICA: Add table-driven dispatch for sleep/wake functions Len Brown
2012-03-30 10:13 ` [PATCH 33/76] ACPICA: Update to version 20120215 Len Brown
2012-03-30 10:13 ` [PATCH 34/76] ACPICA: Clarify METHOD_NAME* defines for full-pathname cases Len Brown
2012-03-30 10:13 ` [PATCH 35/76] ACPICA: Change exception code for invalid pathname in acpi_evaluate_object Len Brown
2012-03-30 10:13 ` [PATCH 36/76] ACPICA: Debugger: Add missing object info to namespace dump Len Brown
2012-03-30 10:13 ` [PATCH 37/76] ACPICA: Sleep/Wake interfaces: optionally execute _GTS and _BFS Len Brown
2012-03-30 10:13 ` [PATCH 38/76] ACPI: Move module parameter gts and bfs to sleep.c Len Brown
2012-03-30 10:13 ` [PATCH 39/76] tools turbostat: add summary option Len Brown
2012-03-30 10:13 ` [PATCH 40/76] tools turbostat: reduce measurement overhead due to IPIs Len Brown
2012-03-30 10:13 ` [PATCH 41/76] tools turbostat: harden against cpu online/offline Len Brown
2012-03-30 10:13 ` [PATCH 42/76] ACPI: ec: Do request_region outside WARN() Len Brown
2012-03-30 10:13 ` [PATCH 43/76] ACPI: Make ACPI interrupt threaded Len Brown
2012-03-30 10:13 ` [PATCH 44/76] ACPICA: Object repair code: Support to add Package wrappers Len Brown
2012-03-30 10:13 ` [PATCH 45/76] ACPICA: Update to version 20120320 Len Brown
2012-03-30 10:13 ` [PATCH 46/76] ACPI: Introduce ACPI D3_COLD state support Len Brown
2012-04-01 6:53 ` [linux-pm] " Rafael J. Wysocki
2012-03-30 10:13 ` [PATCH 47/76] ACPI: Add interface to register/unregister device to/from power resources Len Brown
2012-03-30 10:13 ` [PATCH 48/76] cpuidle: add a sysfs entry to disable specific C state for debug purpose Len Brown
2012-03-30 10:13 ` [PATCH 49/76] cpuidle: use the driver's state_count as default Len Brown
2012-03-30 10:13 ` [PATCH 50/76] cpuidle: remove useless array definition in cpuidle_structure Len Brown
2012-03-30 10:13 ` [PATCH 51/76] cpuidle: remove unused 'governor_data' field Len Brown
2012-03-30 10:13 ` [PATCH 52/76] ACPI, PCI: Move acpi_dev_run_wake() to ACPI core Len Brown
2012-03-30 10:13 ` [PATCH 53/76] ACPI: Evaluate thermal trip points before reading temperature Len Brown
2012-03-30 10:13 ` [PATCH 54/76] ACPI: Ensure thermal limits match CPU frequencies Len Brown
2012-03-30 10:13 ` [PATCH 55/76] ACPI / PM: print physical addresses consistently with other parts of kernel Len Brown
2012-03-30 10:13 ` [PATCH 56/76] ACPI: Add CPU hotplug support for processor device objects Len Brown
2012-03-30 10:14 ` [PATCH 57/76] ACPI / Video: blacklist some samsung laptops Len Brown
2012-03-30 12:07 ` Corentin Chary
2012-03-30 12:16 ` Len Brown
2012-03-30 10:14 ` Len Brown [this message]
2012-04-02 16:13 ` [PATCH 58/76] idle, x86: Allow off-lined CPU to enter deeper C states Tony Luck
2012-04-02 17:25 ` Tony Luck
2012-04-02 17:45 ` Konrad Rzeszutek Wilk
2012-04-02 17:56 ` Boris Ostrovsky
2012-04-02 18:02 ` Tony Luck
2012-04-02 18:10 ` Boris Ostrovsky
2012-03-30 10:14 ` [PATCH 59/76] cpuidle: power_usage should be declared signed integer Len Brown
2012-03-30 10:14 ` [PATCH 60/76] ACPI, APEI, Fix ERST header length check Len Brown
2012-03-30 10:14 ` [PATCH 61/76] ACPI, APEI, EINJ, limit the range of einj_param Len Brown
2012-03-30 10:14 ` [PATCH 62/76] ACPI, APEI, EINJ, new parameter to control trigger action Len Brown
2012-03-30 10:14 ` [PATCH 63/76] Update documentation for parameter *notrigger* in einj.txt Len Brown
2012-03-30 10:14 ` [PATCH 64/76] ACPI, APEI: Fix incorrect APEI register bit width check and usage Len Brown
2012-03-30 10:14 ` [PATCH 65/76] ACPI: processor_driver: add missing kfree Len Brown
2012-03-30 10:14 ` [PATCH 66/76] ACPI: Fix use-after-free in acpi_map_lsapic Len Brown
2012-03-30 10:14 ` [PATCH 67/76] PNPACPI: Fix device ref leaking in acpi_pnp_match Len Brown
2012-03-30 10:14 ` [PATCH 68/76] ACPI: consistently use should_use_kmap() Len Brown
2012-03-30 10:14 ` [PATCH 69/76] ACPI: Fix unprotected smp_processor_id() in acpi_processor_cst_has_changed() Len Brown
2012-03-30 10:14 ` [PATCH 70/76] ACPI: Clean redundant codes in scan.c Len Brown
2012-03-30 10:14 ` [PATCH 71/76] CPER failed to handle generic error records with multiple sections Len Brown
2012-03-30 10:14 ` [PATCH 72/76] ACPI: Fix logic for removing mappings in 'acpi_unmap' Len Brown
2012-03-30 10:14 ` [PATCH 73/76] ACPI: export acpi_kobj Len Brown
2012-03-30 10:14 ` [PATCH 74/76] ACPI: Add support for exposing BGRT data Len Brown
2012-03-30 10:14 ` [PATCH 75/76] Disable MCP limit exceeded messages from Intel IPS driver Len Brown
2012-03-30 10:14 ` [PATCH 76/76] ACPI throttling: fix endian bug in acpi_read_throttling_status() Len Brown
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1a022e3f1be11730bd8747b1af96a0274bf6356e.1333101989.git.len.brown@intel.com \
--to=lenb@kernel.org \
--cc=boris.ostrovsky@amd.com \
--cc=len.brown@intel.com \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@lists.linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®