* [PATCH 62/76] ACPI, APEI, EINJ, new parameter to control trigger action
From: Len Brown @ 2012-03-30 10:14 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Chen Gong, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Chen Gong <gong.chen@linux.intel.com>
Some APEI firmware implementation will access injected address
specified in param1 to trigger the error when injecting memory
error, which means if one SRAR error is injected, the crash
always happens because it is executed in kernel context. This
new parameter can disable trigger action and control is taken
over by the user. In this way, an SRAR error can happen in user
context instead of crashing the system. This function is highly
depended on BIOS implementation so please ensure you know the
BIOS trigger procedure before you enable this switch.
v2:
notrigger should be created together with param1/param2
Tested-by: Tony Luck <tony.luck@lintel.com>
Signed-off-by: Chen Gong <gong.chen@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/apei/einj.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/drivers/acpi/apei/einj.c b/drivers/acpi/apei/einj.c
index 3d3816f..8e17936 100644
--- a/drivers/acpi/apei/einj.c
+++ b/drivers/acpi/apei/einj.c
@@ -74,6 +74,8 @@ struct vendor_error_type_extension {
u8 reserved[3];
};
+static u32 notrigger;
+
static u32 vendor_flags;
static struct debugfs_blob_wrapper vendor_blob;
static char vendor_dev[64];
@@ -496,9 +498,11 @@ static int __einj_error_inject(u32 type, u64 param1, u64 param2)
if (rc)
return rc;
trigger_paddr = apei_exec_ctx_get_output(&ctx);
- rc = __einj_error_trigger(trigger_paddr, type, param1, param2);
- if (rc)
- return rc;
+ if (notrigger == 0) {
+ rc = __einj_error_trigger(trigger_paddr, type, param1, param2);
+ if (rc)
+ return rc;
+ }
rc = apei_exec_run_optional(&ctx, ACPI_EINJ_END_OPERATION);
return rc;
@@ -700,6 +704,11 @@ static int __init einj_init(void)
einj_debug_dir, &error_param2);
if (!fentry)
goto err_unmap;
+
+ fentry = debugfs_create_x32("notrigger", S_IRUSR | S_IWUSR,
+ einj_debug_dir, ¬rigger);
+ if (!fentry)
+ goto err_unmap;
}
if (vendor_dev[0]) {
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 61/76] ACPI, APEI, EINJ, limit the range of einj_param
From: Len Brown @ 2012-03-30 10:14 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Chen Gong, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Chen Gong <gong.chen@linux.intel.com>
On the platforms with ACPI4.x support, parameter extension
is not always doable, which means only parameter extension
is enabled, einj_param can take effect.
v2->v1: stopping early in einj_get_parameter_address for einj_param
Signed-off-by: Chen Gong <gong.chen@linux.intel.com>
Acked-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/apei/einj.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/acpi/apei/einj.c b/drivers/acpi/apei/einj.c
index 4ca087d..3d3816f 100644
--- a/drivers/acpi/apei/einj.c
+++ b/drivers/acpi/apei/einj.c
@@ -238,7 +238,7 @@ static void *einj_get_parameter_address(void)
return v5param;
}
}
- if (paddrv4) {
+ if (param_extension && paddrv4) {
struct einj_parameter *v4param;
v4param = acpi_os_map_memory(paddrv4, sizeof(*v4param));
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 60/76] ACPI, APEI, Fix ERST header length check
From: Len Brown @ 2012-03-30 10:14 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Len Brown, linux-kernel, Jiang Liu, Jiang Liu
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Jiang Liu <liuj97@gmail.com>
This fixes a trivial copy & paste error in ERST header length check.
It's just for future safety because sizeof(struct acpi_table_einj)
equals to sizeof(struct acpi_table_erst) with current ACPI5.0
specification. It applies to v3.3-rc6.
Signed-off-by: Jiang Liu <jiang.liu@huawei.com>
Acked-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/apei/erst.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/acpi/apei/erst.c b/drivers/acpi/apei/erst.c
index eb9fab5..e4d9d24 100644
--- a/drivers/acpi/apei/erst.c
+++ b/drivers/acpi/apei/erst.c
@@ -917,7 +917,7 @@ static int erst_check_table(struct acpi_table_erst *erst_tab)
{
if ((erst_tab->header_length !=
(sizeof(struct acpi_table_erst) - sizeof(erst_tab->header)))
- && (erst_tab->header_length != sizeof(struct acpi_table_einj)))
+ && (erst_tab->header_length != sizeof(struct acpi_table_erst)))
return -EINVAL;
if (erst_tab->header.length < sizeof(struct acpi_table_erst))
return -EINVAL;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 59/76] cpuidle: power_usage should be declared signed integer
From: Len Brown @ 2012-03-30 10:14 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Boris Ostrovsky, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Boris Ostrovsky <boris.ostrovsky@amd.com>
power_usage is always assigned a negative value and should be declared
a signed integer
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@amd.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
| 2 +-
include/linux/cpuidle.h | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
--git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index 5c17ca1..0633575 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -236,7 +236,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev)
{
struct menu_device *data = &__get_cpu_var(menu_devices);
int latency_req = pm_qos_request(PM_QOS_CPU_DMA_LATENCY);
- unsigned int power_usage = -1;
+ int power_usage = -1;
int i;
int multiplier;
struct timespec t;
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index d557bcd..6c26a3d 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -44,7 +44,7 @@ struct cpuidle_state {
unsigned int flags;
unsigned int exit_latency; /* in US */
- unsigned int power_usage; /* in mW */
+ int power_usage; /* in mW */
unsigned int target_residency; /* in US */
unsigned int disable;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 58/76] idle, x86: Allow off-lined CPU to enter deeper C states
From: Len Brown @ 2012-03-30 10:14 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Boris Ostrovsky, linux-kernel, Len Brown
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
^ permalink raw reply related
* [PATCH 57/76] ACPI / Video: blacklist some samsung laptops
From: Len Brown @ 2012-03-30 10:14 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Corentin Chary, Len Brown, linux-kernel
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Corentin Chary <corentincj@iksaif.net>
On these laptops, the ACPI video is not functional, and very unlikely
to be fixed by the vendor. Note that intel_backlight works for some
of these laptops, and the backlight from samsung-laptop always work.
The good news is that newer laptops have functional ACPI video device
and won't end up growing this list.
Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/video_detect.c | 49 +++++++++++++++++++++++++++++++++++++++++++
1 file changed, 49 insertions(+)
diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c
index f3f0fe7..b6f6979 100644
--- a/drivers/acpi/video_detect.c
+++ b/drivers/acpi/video_detect.c
@@ -132,6 +132,53 @@ find_video(acpi_handle handle, u32 lvl, void *context, void **rv)
return AE_OK;
}
+/* Force to use vendor driver when the ACPI device is known to be
+ * buggy */
+static int video_detect_force_vendor(const struct dmi_system_id *d)
+{
+ acpi_video_support |= ACPI_VIDEO_BACKLIGHT_DMI_VENDOR;
+ return 0;
+}
+
+static struct dmi_system_id video_detect_dmi_table[] = {
+ {
+ .callback = video_detect_force_vendor,
+ .ident = "N150P",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "N150P"),
+ DMI_MATCH(DMI_BOARD_NAME, "N150P"),
+ },
+ },
+ {
+ .callback = video_detect_force_vendor,
+ .ident = "N145P/N250P/N260P",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "N145P/N250P/N260P"),
+ DMI_MATCH(DMI_BOARD_NAME, "N145P/N250P/N260P"),
+ },
+ },
+ {
+ .callback = video_detect_force_vendor,
+ .ident = "N150/N210/N220",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "N150/N210/N220"),
+ DMI_MATCH(DMI_BOARD_NAME, "N150/N210/N220"),
+ },
+ },
+ {
+ .callback = video_detect_force_vendor,
+ .ident = "NF110/NF210/NF310",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "NF110/NF210/NF310"),
+ DMI_MATCH(DMI_BOARD_NAME, "NF110/NF210/NF310"),
+ },
+ },
+};
+
/*
* Returns the video capabilities of a specific ACPI graphics device
*
@@ -164,6 +211,8 @@ long acpi_video_get_capabilities(acpi_handle graphics_handle)
* ACPI_VIDEO_BACKLIGHT_DMI_VENDOR;
*}
*/
+
+ dmi_check_system(video_detect_dmi_table);
} else {
status = acpi_bus_get_device(graphics_handle, &tmp_dev);
if (ACPI_FAILURE(status)) {
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 56/76] ACPI: Add CPU hotplug support for processor device objects
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Len Brown, Toshi Kani, linux-kernel
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Toshi Kani <toshi.kani@hp.com>
acpi_processor_install_hotplug_notify() registers processor objects to
receive ACPI CPU hotplug event notifications. This patch additionally
registers processor device objects (ACPI0007) to receive the notifications
as well.
Signed-off-by: Toshi Kani <toshi.kani@hp.com>
Reviewed-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/processor_driver.c | 48 +++++++++++++++++++++++++++++++--------
1 file changed, 38 insertions(+), 10 deletions(-)
diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c
index 8ae05ce..50be277 100644
--- a/drivers/acpi/processor_driver.c
+++ b/drivers/acpi/processor_driver.c
@@ -68,6 +68,7 @@
#define ACPI_PROCESSOR_NOTIFY_PERFORMANCE 0x80
#define ACPI_PROCESSOR_NOTIFY_POWER 0x81
#define ACPI_PROCESSOR_NOTIFY_THROTTLING 0x82
+#define ACPI_PROCESSOR_DEVICE_HID "ACPI0007"
#define ACPI_PROCESSOR_LIMIT_USER 0
#define ACPI_PROCESSOR_LIMIT_THERMAL 1
@@ -88,7 +89,7 @@ static int acpi_processor_start(struct acpi_processor *pr);
static const struct acpi_device_id processor_device_ids[] = {
{ACPI_PROCESSOR_OBJECT_HID, 0},
- {"ACPI0007", 0},
+ {ACPI_PROCESSOR_DEVICE_HID, 0},
{"", 0},
};
MODULE_DEVICE_TABLE(acpi, processor_device_ids);
@@ -741,20 +742,46 @@ static void acpi_processor_hotplug_notify(acpi_handle handle,
return;
}
+static acpi_status is_processor_device(acpi_handle handle)
+{
+ struct acpi_device_info *info;
+ char *hid;
+ acpi_status status;
+
+ status = acpi_get_object_info(handle, &info);
+ if (ACPI_FAILURE(status))
+ return status;
+
+ if (info->type == ACPI_TYPE_PROCESSOR) {
+ kfree(info);
+ return AE_OK; /* found a processor object */
+ }
+
+ if (!(info->valid & ACPI_VALID_HID)) {
+ kfree(info);
+ return AE_ERROR;
+ }
+
+ hid = info->hardware_id.string;
+ if ((hid == NULL) || strcmp(hid, ACPI_PROCESSOR_DEVICE_HID)) {
+ kfree(info);
+ return AE_ERROR;
+ }
+
+ kfree(info);
+ return AE_OK; /* found a processor device object */
+}
+
static acpi_status
processor_walk_namespace_cb(acpi_handle handle,
u32 lvl, void *context, void **rv)
{
acpi_status status;
int *action = context;
- acpi_object_type type = 0;
- status = acpi_get_type(handle, &type);
+ status = is_processor_device(handle);
if (ACPI_FAILURE(status))
- return (AE_OK);
-
- if (type != ACPI_TYPE_PROCESSOR)
- return (AE_OK);
+ return AE_OK; /* not a processor; continue to walk */
switch (*action) {
case INSTALL_NOTIFY_HANDLER:
@@ -772,7 +799,8 @@ processor_walk_namespace_cb(acpi_handle handle,
break;
}
- return (AE_OK);
+ /* found a processor; skip walking underneath */
+ return AE_CTRL_DEPTH;
}
static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)
@@ -830,7 +858,7 @@ void acpi_processor_install_hotplug_notify(void)
{
#ifdef CONFIG_ACPI_HOTPLUG_CPU
int action = INSTALL_NOTIFY_HANDLER;
- acpi_walk_namespace(ACPI_TYPE_PROCESSOR,
+ acpi_walk_namespace(ACPI_TYPE_ANY,
ACPI_ROOT_OBJECT,
ACPI_UINT32_MAX,
processor_walk_namespace_cb, NULL, &action, NULL);
@@ -843,7 +871,7 @@ void acpi_processor_uninstall_hotplug_notify(void)
{
#ifdef CONFIG_ACPI_HOTPLUG_CPU
int action = UNINSTALL_NOTIFY_HANDLER;
- acpi_walk_namespace(ACPI_TYPE_PROCESSOR,
+ acpi_walk_namespace(ACPI_TYPE_ANY,
ACPI_ROOT_OBJECT,
ACPI_UINT32_MAX,
processor_walk_namespace_cb, NULL, &action, NULL);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 55/76] ACPI / PM: print physical addresses consistently with other parts of kernel
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Bjorn Helgaas, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Bjorn Helgaas <bhelgaas@google.com>
Print physical address info in a style consistent with the %pR style used
elsewhere in the kernel.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/nvs.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/acpi/nvs.c b/drivers/acpi/nvs.c
index 7a2035f..266bc58 100644
--- a/drivers/acpi/nvs.c
+++ b/drivers/acpi/nvs.c
@@ -95,8 +95,8 @@ static int suspend_nvs_register(unsigned long start, unsigned long size)
{
struct nvs_page *entry, *next;
- pr_info("PM: Registering ACPI NVS region at %lx (%ld bytes)\n",
- start, size);
+ pr_info("PM: Registering ACPI NVS region [mem %#010lx-%#010lx] (%ld bytes)\n",
+ start, start + size - 1, size);
while (size > 0) {
unsigned int nr_bytes;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 54/76] ACPI: Ensure thermal limits match CPU frequencies
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Matthew Garrett, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Matthew Garrett <mjg@redhat.com>
The ACPI thermal management code supports slowing down a CPU when it's
overheating. Right now that's done by choosing to run it at 100%, 75%, 50%
or 25% of full speed. However, most CPUs do not allow an arbitrary
frequency to be set and so will run at the first frequency below that value.
This doesn't match the intent of the specification, which is to drop the
frequency state by state until the temperature stabalises. Fix this up
so it uses actual frequencies rather than percentages.
Reported by: Gene Snider <snider6982@comcast.net>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/processor_thermal.c | 45 +++++++++++++++++++++-----------------
1 file changed, 25 insertions(+), 20 deletions(-)
diff --git a/drivers/acpi/processor_thermal.c b/drivers/acpi/processor_thermal.c
index 3b599ab..1fbf539 100644
--- a/drivers/acpi/processor_thermal.c
+++ b/drivers/acpi/processor_thermal.c
@@ -51,10 +51,8 @@ ACPI_MODULE_NAME("processor_thermal");
* _any_ cpufreq driver and not only the acpi-cpufreq driver.
*/
-#define CPUFREQ_THERMAL_MIN_STEP 0
-#define CPUFREQ_THERMAL_MAX_STEP 3
-static DEFINE_PER_CPU(unsigned int, cpufreq_thermal_reduction_pctg);
+static DEFINE_PER_CPU(unsigned int, cpufreq_thermal_limit_state);
static unsigned int acpi_thermal_cpufreq_is_init = 0;
static int cpu_has_cpufreq(unsigned int cpu)
@@ -69,19 +67,19 @@ static int acpi_thermal_cpufreq_notifier(struct notifier_block *nb,
unsigned long event, void *data)
{
struct cpufreq_policy *policy = data;
- unsigned long max_freq = 0;
+ int state = per_cpu(cpufreq_thermal_limit_state, policy->cpu);
+ struct cpufreq_frequency_table *table;
if (event != CPUFREQ_ADJUST)
- goto out;
+ return 0;
+
+ table = cpufreq_frequency_get_table(policy->cpu);
- max_freq = (
- policy->cpuinfo.max_freq *
- (100 - per_cpu(cpufreq_thermal_reduction_pctg, policy->cpu) * 20)
- ) / 100;
+ if (!table)
+ return 0;
- cpufreq_verify_within_limits(policy, 0, max_freq);
+ cpufreq_verify_within_limits(policy, 0, table[state].frequency);
- out:
return 0;
}
@@ -91,10 +89,21 @@ static struct notifier_block acpi_thermal_cpufreq_notifier_block = {
static int cpufreq_get_max_state(unsigned int cpu)
{
+ int count = 0;
+ struct cpufreq_frequency_table *table;
+
if (!cpu_has_cpufreq(cpu))
return 0;
- return CPUFREQ_THERMAL_MAX_STEP;
+ table = cpufreq_frequency_get_table(cpu);
+
+ if (!table)
+ return 0;
+
+ while (table[count].frequency != CPUFREQ_TABLE_END)
+ count++;
+
+ return count;
}
static int cpufreq_get_cur_state(unsigned int cpu)
@@ -102,7 +111,7 @@ static int cpufreq_get_cur_state(unsigned int cpu)
if (!cpu_has_cpufreq(cpu))
return 0;
- return per_cpu(cpufreq_thermal_reduction_pctg, cpu);
+ return per_cpu(cpufreq_thermal_limit_state, cpu);
}
static int cpufreq_set_cur_state(unsigned int cpu, int state)
@@ -110,7 +119,7 @@ static int cpufreq_set_cur_state(unsigned int cpu, int state)
if (!cpu_has_cpufreq(cpu))
return 0;
- per_cpu(cpufreq_thermal_reduction_pctg, cpu) = state;
+ per_cpu(cpufreq_thermal_limit_state, cpu) = state;
cpufreq_update_policy(cpu);
return 0;
}
@@ -121,7 +130,7 @@ void acpi_thermal_cpufreq_init(void)
for (i = 0; i < nr_cpu_ids; i++)
if (cpu_present(i))
- per_cpu(cpufreq_thermal_reduction_pctg, i) = 0;
+ per_cpu(cpufreq_thermal_limit_state, i) = 0;
i = cpufreq_register_notifier(&acpi_thermal_cpufreq_notifier_block,
CPUFREQ_POLICY_NOTIFIER);
@@ -169,15 +178,11 @@ int acpi_processor_get_limit_info(struct acpi_processor *pr)
return 0;
}
-/* thermal coolign device callbacks */
+/* thermal cooling device callbacks */
static int acpi_processor_max_state(struct acpi_processor *pr)
{
int max_state = 0;
- /*
- * There exists four states according to
- * cpufreq_thermal_reduction_ptg. 0, 1, 2, 3
- */
max_state += cpufreq_get_max_state(pr->id);
if (pr->flags.throttling)
max_state += (pr->throttling.state_count -1);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 53/76] ACPI: Evaluate thermal trip points before reading temperature
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Matthew Garrett, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Matthew Garrett <mjg@redhat.com>
An HP laptop (Pavilion G4-1016tx) has the following code in _TMP:
Store (\_SB.PCI0.LPCB.EC0.RTMP, Local0)
If (LGreaterEqual (Local0, S4TP))
{
Store (One, HTS4)
}
S4TP is initialised at 0 and not programmed further until either _HOT or
_CRT is called. If we evaluate _TMP before the trip points then HTS4 will
always be set, causing the firmware to generate a message on boot
complaining that the system shut down because of overheating. The simplest
solution is just to reverse the checking of trip points and _TMP in thermal
init.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/thermal.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c
index 48fbc64..7dbebea 100644
--- a/drivers/acpi/thermal.c
+++ b/drivers/acpi/thermal.c
@@ -941,13 +941,13 @@ static int acpi_thermal_get_info(struct acpi_thermal *tz)
if (!tz)
return -EINVAL;
- /* Get temperature [_TMP] (required) */
- result = acpi_thermal_get_temperature(tz);
+ /* Get trip points [_CRT, _PSV, etc.] (required) */
+ result = acpi_thermal_get_trip_points(tz);
if (result)
return result;
- /* Get trip points [_CRT, _PSV, etc.] (required) */
- result = acpi_thermal_get_trip_points(tz);
+ /* Get temperature [_TMP] (required) */
+ result = acpi_thermal_get_temperature(tz);
if (result)
return result;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 52/76] ACPI, PCI: Move acpi_dev_run_wake() to ACPI core
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Len Brown, linux-kernel, Lin Ming
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Lin Ming <ming.m.lin@intel.com>
acpi_dev_run_wake() is a generic function which can be used by
other subsystem too. Rename it to acpi_pm_device_run_wake, to be
consistent with acpi_pm_device_sleep_wake.
Then move it to ACPI core.
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/sleep.c | 35 +++++++++++++++++++++++++++++++++++
drivers/pci/pci-acpi.c | 40 +++-------------------------------------
include/acpi/acpi_bus.h | 5 +++++
3 files changed, 43 insertions(+), 37 deletions(-)
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index ca191ff..00fa664 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -17,6 +17,7 @@
#include <linux/suspend.h>
#include <linux/reboot.h>
#include <linux/acpi.h>
+#include <linux/pm_runtime.h>
#include <asm/io.h>
@@ -730,6 +731,40 @@ int acpi_pm_device_sleep_state(struct device *dev, int *d_min_p)
#ifdef CONFIG_PM_SLEEP
/**
+ * acpi_pm_device_run_wake - Enable/disable wake-up for given device.
+ * @phys_dev: Device to enable/disable the platform to wake-up the system for.
+ * @enable: Whether enable or disable the wake-up functionality.
+ *
+ * Find the ACPI device object corresponding to @pci_dev and try to
+ * enable/disable the GPE associated with it.
+ */
+int acpi_pm_device_run_wake(struct device *phys_dev, bool enable)
+{
+ struct acpi_device *dev;
+ acpi_handle handle;
+
+ if (!device_run_wake(phys_dev))
+ return -EINVAL;
+
+ handle = DEVICE_ACPI_HANDLE(phys_dev);
+ if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &dev))) {
+ dev_dbg(phys_dev, "ACPI handle has no context in %s!\n",
+ __func__);
+ return -ENODEV;
+ }
+
+ if (enable) {
+ acpi_enable_wakeup_device_power(dev, ACPI_STATE_S0);
+ acpi_enable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number);
+ } else {
+ acpi_disable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number);
+ acpi_disable_wakeup_device_power(dev);
+ }
+
+ return 0;
+}
+
+/**
* acpi_pm_device_sleep_wake - enable or disable the system wake-up
* capability of given device
* @dev: device to handle
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 060fd22..0f150f2 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -277,40 +277,6 @@ static int acpi_pci_sleep_wake(struct pci_dev *dev, bool enable)
return 0;
}
-/**
- * acpi_dev_run_wake - Enable/disable wake-up for given device.
- * @phys_dev: Device to enable/disable the platform to wake-up the system for.
- * @enable: Whether enable or disable the wake-up functionality.
- *
- * Find the ACPI device object corresponding to @pci_dev and try to
- * enable/disable the GPE associated with it.
- */
-static int acpi_dev_run_wake(struct device *phys_dev, bool enable)
-{
- struct acpi_device *dev;
- acpi_handle handle;
-
- if (!device_run_wake(phys_dev))
- return -EINVAL;
-
- handle = DEVICE_ACPI_HANDLE(phys_dev);
- if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &dev))) {
- dev_dbg(phys_dev, "ACPI handle has no context in %s!\n",
- __func__);
- return -ENODEV;
- }
-
- if (enable) {
- acpi_enable_wakeup_device_power(dev, ACPI_STATE_S0);
- acpi_enable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number);
- } else {
- acpi_disable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number);
- acpi_disable_wakeup_device_power(dev);
- }
-
- return 0;
-}
-
static void acpi_pci_propagate_run_wake(struct pci_bus *bus, bool enable)
{
while (bus->parent) {
@@ -318,14 +284,14 @@ static void acpi_pci_propagate_run_wake(struct pci_bus *bus, bool enable)
if (bridge->pme_interrupt)
return;
- if (!acpi_dev_run_wake(&bridge->dev, enable))
+ if (!acpi_pm_device_run_wake(&bridge->dev, enable))
return;
bus = bus->parent;
}
/* We have reached the root bus. */
if (bus->bridge)
- acpi_dev_run_wake(bus->bridge, enable);
+ acpi_pm_device_run_wake(bus->bridge, enable);
}
static int acpi_pci_run_wake(struct pci_dev *dev, bool enable)
@@ -333,7 +299,7 @@ static int acpi_pci_run_wake(struct pci_dev *dev, bool enable)
if (dev->pme_interrupt)
return 0;
- if (!acpi_dev_run_wake(&dev->dev, enable))
+ if (!acpi_pm_device_run_wake(&dev->dev, enable))
return 0;
acpi_pci_propagate_run_wake(dev->bus, enable);
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index e168fff..f1c8ca6 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -394,8 +394,13 @@ static inline int acpi_pm_device_sleep_state(struct device *d, int *p)
#endif
#ifdef CONFIG_PM_SLEEP
+int acpi_pm_device_run_wake(struct device *, bool);
int acpi_pm_device_sleep_wake(struct device *, bool);
#else
+static inline int acpi_pm_device_run_wake(struct device *dev, bool enable)
+{
+ return -ENODEV;
+}
static inline int acpi_pm_device_sleep_wake(struct device *dev, bool enable)
{
return -ENODEV;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 51/76] cpuidle: remove unused 'governor_data' field
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Daniel Lezcano, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Daniel Lezcano <daniel.lezcano@linaro.org>
As far as I can see, this field is never used in the code.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
include/linux/cpuidle.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index f7f1d90..f3ebbba 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -98,7 +98,6 @@ struct cpuidle_device {
struct list_head device_list;
struct kobject kobj;
struct completion kobj_unregister;
- void *governor_data;
};
DECLARE_PER_CPU(struct cpuidle_device *, cpuidle_devices);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 50/76] cpuidle: remove useless array definition in cpuidle_structure
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Daniel Lezcano, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Daniel Lezcano <daniel.lezcano@linaro.org>
All the modules name are ro-data, it is never copied to the array.
eg.
static struct cpuidle_driver intel_idle_driver = {
.name = "intel_idle",
.owner = THIS_MODULE,
};
It safe to assign the pointer of this ro-data to a const char *.
By this way we save 12 bytes.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Tested-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
include/linux/cpuidle.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index ca4e498..f7f1d90 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -120,7 +120,7 @@ static inline int cpuidle_get_last_residency(struct cpuidle_device *dev)
****************************/
struct cpuidle_driver {
- char name[CPUIDLE_NAME_LEN];
+ const char *name;
struct module *owner;
unsigned int power_specified:1;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 49/76] cpuidle: use the driver's state_count as default
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Daniel Lezcano, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Daniel Lezcano <daniel.lezcano@linaro.org>
If the state_count is not initialized for the device use
the driver's state count as the default. That will prevent
to add it manually in the cpuidle driver initialization
routine and will save us from duplicate line of code.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/cpuidle/cpuidle.c | 2 +-
drivers/cpuidle/driver.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 77304b6..f7cab5e 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -268,7 +268,7 @@ int cpuidle_enable_device(struct cpuidle_device *dev)
if (!drv || !cpuidle_curr_governor)
return -EIO;
if (!dev->state_count)
- return -EINVAL;
+ dev->state_count = drv->state_count;
if (dev->registered == 0) {
ret = __cpuidle_register_device(dev);
diff --git a/drivers/cpuidle/driver.c b/drivers/cpuidle/driver.c
index 284d7af..40cd3f3 100644
--- a/drivers/cpuidle/driver.c
+++ b/drivers/cpuidle/driver.c
@@ -47,7 +47,7 @@ static void __cpuidle_register_driver(struct cpuidle_driver *drv)
*/
int cpuidle_register_driver(struct cpuidle_driver *drv)
{
- if (!drv)
+ if (!drv || !drv->state_count)
return -EINVAL;
if (cpuidle_disabled())
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 48/76] cpuidle: add a sysfs entry to disable specific C state for debug purpose.
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: ShuoX Liu, Andrew Morton, linux-kernel, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: ShuoX Liu <shuox.liu@intel.com>
Some C states of new CPU might be not good. One reason is BIOS might
configure them incorrectly. To help developers root cause it quickly, the
patch adds a new sysfs entry, so developers could disable specific C state
manually.
In addition, C state might have much impact on performance tuning, as it
takes much time to enter/exit C states, which might delay interrupt
processing. With the new debug option, developers could check if a deep C
state could impact performance and how much impact it could cause.
Also add this option in Documentation/cpuidle/sysfs.txt.
[akpm@linux-foundation.org: check kstrtol return value]
Signed-off-by: ShuoX Liu <shuox.liu@intel.com>
Reviewed-by: Yanmin Zhang <yanmin_zhang@intel.com>
Reviewed-and-Tested-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
Documentation/cpuidle/sysfs.txt | 5 +++++
drivers/cpuidle/cpuidle.c | 1 +
| 5 ++++-
drivers/cpuidle/sysfs.c | 40 ++++++++++++++++++++++++++++++++++++++
include/linux/cpuidle.h | 1 +
5 files changed, 51 insertions(+), 1 deletion(-)
diff --git a/Documentation/cpuidle/sysfs.txt b/Documentation/cpuidle/sysfs.txt
index 50d7b16..9d28a34 100644
--- a/Documentation/cpuidle/sysfs.txt
+++ b/Documentation/cpuidle/sysfs.txt
@@ -36,6 +36,7 @@ drwxr-xr-x 2 root root 0 Feb 8 10:42 state3
/sys/devices/system/cpu/cpu0/cpuidle/state0:
total 0
-r--r--r-- 1 root root 4096 Feb 8 10:42 desc
+-rw-r--r-- 1 root root 4096 Feb 8 10:42 disable
-r--r--r-- 1 root root 4096 Feb 8 10:42 latency
-r--r--r-- 1 root root 4096 Feb 8 10:42 name
-r--r--r-- 1 root root 4096 Feb 8 10:42 power
@@ -45,6 +46,7 @@ total 0
/sys/devices/system/cpu/cpu0/cpuidle/state1:
total 0
-r--r--r-- 1 root root 4096 Feb 8 10:42 desc
+-rw-r--r-- 1 root root 4096 Feb 8 10:42 disable
-r--r--r-- 1 root root 4096 Feb 8 10:42 latency
-r--r--r-- 1 root root 4096 Feb 8 10:42 name
-r--r--r-- 1 root root 4096 Feb 8 10:42 power
@@ -54,6 +56,7 @@ total 0
/sys/devices/system/cpu/cpu0/cpuidle/state2:
total 0
-r--r--r-- 1 root root 4096 Feb 8 10:42 desc
+-rw-r--r-- 1 root root 4096 Feb 8 10:42 disable
-r--r--r-- 1 root root 4096 Feb 8 10:42 latency
-r--r--r-- 1 root root 4096 Feb 8 10:42 name
-r--r--r-- 1 root root 4096 Feb 8 10:42 power
@@ -63,6 +66,7 @@ total 0
/sys/devices/system/cpu/cpu0/cpuidle/state3:
total 0
-r--r--r-- 1 root root 4096 Feb 8 10:42 desc
+-rw-r--r-- 1 root root 4096 Feb 8 10:42 disable
-r--r--r-- 1 root root 4096 Feb 8 10:42 latency
-r--r--r-- 1 root root 4096 Feb 8 10:42 name
-r--r--r-- 1 root root 4096 Feb 8 10:42 power
@@ -72,6 +76,7 @@ total 0
* desc : Small description about the idle state (string)
+* disable : Option to disable this idle state (bool)
* latency : Latency to exit out of this idle state (in microseconds)
* name : Name of the idle state (string)
* power : Power consumed while in this idle state (in milliwatts)
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 4869b55..77304b6 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -245,6 +245,7 @@ static void poll_idle_init(struct cpuidle_driver *drv)
state->power_usage = -1;
state->flags = 0;
state->enter = poll_idle;
+ state->disable = 0;
}
#else
static void poll_idle_init(struct cpuidle_driver *drv) {}
--git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index ad09526..5c17ca1 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -280,7 +280,8 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev)
* We want to default to C1 (hlt), not to busy polling
* unless the timer is happening really really soon.
*/
- if (data->expected_us > 5)
+ if (data->expected_us > 5 &&
+ drv->states[CPUIDLE_DRIVER_STATE_START].disable == 0)
data->last_state_idx = CPUIDLE_DRIVER_STATE_START;
/*
@@ -290,6 +291,8 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev)
for (i = CPUIDLE_DRIVER_STATE_START; i < drv->state_count; i++) {
struct cpuidle_state *s = &drv->states[i];
+ if (s->disable)
+ continue;
if (s->target_residency > data->predicted_us)
continue;
if (s->exit_latency > latency_req)
diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
index 3fe41fe..88032b4 100644
--- a/drivers/cpuidle/sysfs.c
+++ b/drivers/cpuidle/sysfs.c
@@ -11,6 +11,7 @@
#include <linux/sysfs.h>
#include <linux/slab.h>
#include <linux/cpu.h>
+#include <linux/capability.h>
#include "cpuidle.h"
@@ -222,6 +223,9 @@ struct cpuidle_state_attr {
#define define_one_state_ro(_name, show) \
static struct cpuidle_state_attr attr_##_name = __ATTR(_name, 0444, show, NULL)
+#define define_one_state_rw(_name, show, store) \
+static struct cpuidle_state_attr attr_##_name = __ATTR(_name, 0644, show, store)
+
#define define_show_state_function(_name) \
static ssize_t show_state_##_name(struct cpuidle_state *state, \
struct cpuidle_state_usage *state_usage, char *buf) \
@@ -229,6 +233,24 @@ static ssize_t show_state_##_name(struct cpuidle_state *state, \
return sprintf(buf, "%u\n", state->_name);\
}
+#define define_store_state_function(_name) \
+static ssize_t store_state_##_name(struct cpuidle_state *state, \
+ const char *buf, size_t size) \
+{ \
+ long value; \
+ int err; \
+ if (!capable(CAP_SYS_ADMIN)) \
+ return -EPERM; \
+ err = kstrtol(buf, 0, &value); \
+ if (err) \
+ return err; \
+ if (value) \
+ state->disable = 1; \
+ else \
+ state->disable = 0; \
+ return size; \
+}
+
#define define_show_state_ull_function(_name) \
static ssize_t show_state_##_name(struct cpuidle_state *state, \
struct cpuidle_state_usage *state_usage, char *buf) \
@@ -251,6 +273,8 @@ define_show_state_ull_function(usage)
define_show_state_ull_function(time)
define_show_state_str_function(name)
define_show_state_str_function(desc)
+define_show_state_function(disable)
+define_store_state_function(disable)
define_one_state_ro(name, show_state_name);
define_one_state_ro(desc, show_state_desc);
@@ -258,6 +282,7 @@ define_one_state_ro(latency, show_state_exit_latency);
define_one_state_ro(power, show_state_power_usage);
define_one_state_ro(usage, show_state_usage);
define_one_state_ro(time, show_state_time);
+define_one_state_rw(disable, show_state_disable, store_state_disable);
static struct attribute *cpuidle_state_default_attrs[] = {
&attr_name.attr,
@@ -266,6 +291,7 @@ static struct attribute *cpuidle_state_default_attrs[] = {
&attr_power.attr,
&attr_usage.attr,
&attr_time.attr,
+ &attr_disable.attr,
NULL
};
@@ -287,8 +313,22 @@ static ssize_t cpuidle_state_show(struct kobject * kobj,
return ret;
}
+static ssize_t cpuidle_state_store(struct kobject *kobj,
+ struct attribute *attr, const char *buf, size_t size)
+{
+ int ret = -EIO;
+ struct cpuidle_state *state = kobj_to_state(kobj);
+ struct cpuidle_state_attr *cattr = attr_to_stateattr(attr);
+
+ if (cattr->store)
+ ret = cattr->store(state, buf, size);
+
+ return ret;
+}
+
static const struct sysfs_ops cpuidle_state_sysfs_ops = {
.show = cpuidle_state_show,
+ .store = cpuidle_state_store,
};
static void cpuidle_state_sysfs_release(struct kobject *kobj)
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 927db28..ca4e498 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -46,6 +46,7 @@ struct cpuidle_state {
unsigned int exit_latency; /* in US */
unsigned int power_usage; /* in mW */
unsigned int target_residency; /* in US */
+ unsigned int disable;
int (*enter) (struct cpuidle_device *dev,
struct cpuidle_driver *drv,
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 47/76] ACPI: Add interface to register/unregister device to/from power resources
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Lin Ming, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Lin Ming <ming.m.lin@intel.com>
Devices may share same list of power resources in _PR0, for example
Device(Dev0)
{
Name (_PR0, Package (0x01)
{
P0PR,
P1PR
})
}
Device(Dev1)
{
Name (_PR0, Package (0x01)
{
P0PR,
P1PR
}
}
Assume Dev0 and Dev1 were runtime suspended.
Then Dev0 is resumed first and it goes into D0 state.
But Dev1 is left in D0_Uninitialised state.
This is wrong. In this case, Dev1 must be resumed too.
In order to hand this case, each power resource maintains a list of
devices which relies on it.
When power resource is ON, it will check if the devices on its list
can be resumed. The device can only be resumed when all the power
resouces of its _PR0 are ON.
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/power.c | 162 +++++++++++++++++++++++++++++++++++++++++++++++
include/acpi/acpi_bus.h | 2 +
2 files changed, 164 insertions(+)
diff --git a/drivers/acpi/power.c b/drivers/acpi/power.c
index 0d681fb..7049a7d 100644
--- a/drivers/acpi/power.c
+++ b/drivers/acpi/power.c
@@ -40,9 +40,11 @@
#include <linux/init.h>
#include <linux/types.h>
#include <linux/slab.h>
+#include <linux/pm_runtime.h>
#include <acpi/acpi_bus.h>
#include <acpi/acpi_drivers.h>
#include "sleep.h"
+#include "internal.h"
#define PREFIX "ACPI: "
@@ -77,6 +79,20 @@ static struct acpi_driver acpi_power_driver = {
},
};
+/*
+ * A power managed device
+ * A device may rely on multiple power resources.
+ * */
+struct acpi_power_managed_device {
+ struct device *dev; /* The physical device */
+ acpi_handle *handle;
+};
+
+struct acpi_power_resource_device {
+ struct acpi_power_managed_device *device;
+ struct acpi_power_resource_device *next;
+};
+
struct acpi_power_resource {
struct acpi_device * device;
acpi_bus_id name;
@@ -84,6 +100,9 @@ struct acpi_power_resource {
u32 order;
unsigned int ref_count;
struct mutex resource_lock;
+
+ /* List of devices relying on this power resource */
+ struct acpi_power_resource_device *devices;
};
static struct list_head acpi_power_resource_list;
@@ -183,8 +202,26 @@ static int acpi_power_get_list_state(struct acpi_handle_list *list, int *state)
return 0;
}
+/* Resume the device when all power resources in _PR0 are on */
+static void acpi_power_on_device(struct acpi_power_managed_device *device)
+{
+ struct acpi_device *acpi_dev;
+ acpi_handle handle = device->handle;
+ int state;
+
+ if (acpi_bus_get_device(handle, &acpi_dev))
+ return;
+
+ if(acpi_power_get_inferred_state(acpi_dev, &state))
+ return;
+
+ if (state == ACPI_STATE_D0 && pm_runtime_suspended(device->dev))
+ pm_request_resume(device->dev);
+}
+
static int __acpi_power_on(struct acpi_power_resource *resource)
{
+ struct acpi_power_resource_device *device_list = resource->devices;
acpi_status status = AE_OK;
status = acpi_evaluate_object(resource->device->handle, "_ON", NULL, NULL);
@@ -197,6 +234,12 @@ static int __acpi_power_on(struct acpi_power_resource *resource)
ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Power resource [%s] turned on\n",
resource->name));
+ while (device_list) {
+ acpi_power_on_device(device_list->device);
+
+ device_list = device_list->next;
+ }
+
return 0;
}
@@ -299,6 +342,125 @@ static int acpi_power_on_list(struct acpi_handle_list *list)
return result;
}
+static void __acpi_power_resource_unregister_device(struct device *dev,
+ acpi_handle res_handle)
+{
+ struct acpi_power_resource *resource = NULL;
+ struct acpi_power_resource_device *prev, *curr;
+
+ if (acpi_power_get_context(res_handle, &resource))
+ return;
+
+ mutex_lock(&resource->resource_lock);
+ prev = NULL;
+ curr = resource->devices;
+ while (curr) {
+ if (curr->device->dev == dev) {
+ if (!prev)
+ resource->devices = curr->next;
+ else
+ prev->next = curr->next;
+
+ kfree(curr);
+ break;
+ }
+
+ prev = curr;
+ curr = curr->next;
+ }
+ mutex_unlock(&resource->resource_lock);
+}
+
+/* Unlink dev from all power resources in _PR0 */
+void acpi_power_resource_unregister_device(struct device *dev, acpi_handle handle)
+{
+ struct acpi_device *acpi_dev;
+ struct acpi_handle_list *list;
+ int i;
+
+ if (!dev || !handle)
+ return;
+
+ if (acpi_bus_get_device(handle, &acpi_dev))
+ return;
+
+ list = &acpi_dev->power.states[ACPI_STATE_D0].resources;
+
+ for (i = 0; i < list->count; i++)
+ __acpi_power_resource_unregister_device(dev,
+ list->handles[i]);
+}
+
+static int __acpi_power_resource_register_device(
+ struct acpi_power_managed_device *powered_device, acpi_handle handle)
+{
+ struct acpi_power_resource *resource = NULL;
+ struct acpi_power_resource_device *power_resource_device;
+ int result;
+
+ result = acpi_power_get_context(handle, &resource);
+ if (result)
+ return result;
+
+ power_resource_device = kzalloc(
+ sizeof(*power_resource_device), GFP_KERNEL);
+ if (!power_resource_device)
+ return -ENOMEM;
+
+ power_resource_device->device = powered_device;
+
+ mutex_lock(&resource->resource_lock);
+ power_resource_device->next = resource->devices;
+ resource->devices = power_resource_device;
+ mutex_unlock(&resource->resource_lock);
+
+ return 0;
+}
+
+/* Link dev to all power resources in _PR0 */
+int acpi_power_resource_register_device(struct device *dev, acpi_handle handle)
+{
+ struct acpi_device *acpi_dev;
+ struct acpi_handle_list *list;
+ struct acpi_power_managed_device *powered_device;
+ int i, ret;
+
+ if (!dev || !handle)
+ return -ENODEV;
+
+ ret = acpi_bus_get_device(handle, &acpi_dev);
+ if (ret)
+ goto no_power_resource;
+
+ if (!acpi_dev->power.flags.power_resources)
+ goto no_power_resource;
+
+ powered_device = kzalloc(sizeof(*powered_device), GFP_KERNEL);
+ if (!powered_device)
+ return -ENOMEM;
+
+ powered_device->dev = dev;
+ powered_device->handle = handle;
+
+ list = &acpi_dev->power.states[ACPI_STATE_D0].resources;
+
+ for (i = 0; i < list->count; i++) {
+ ret = __acpi_power_resource_register_device(powered_device,
+ list->handles[i]);
+
+ if (ret) {
+ acpi_power_resource_unregister_device(dev, handle);
+ break;
+ }
+ }
+
+ return ret;
+
+no_power_resource:
+ printk(KERN_WARNING PREFIX "Invalid Power Resource to register!");
+ return -ENODEV;
+}
+
/**
* acpi_device_sleep_wake - execute _DSW (Device Sleep Wake) or (deprecated in
* ACPI 3.0) _PSW (Power State Wake)
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 6cd5b64..e168fff 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -323,6 +323,8 @@ int acpi_bus_set_power(acpi_handle handle, int state);
int acpi_bus_update_power(acpi_handle handle, int *state_p);
bool acpi_bus_power_manageable(acpi_handle handle);
bool acpi_bus_can_wakeup(acpi_handle handle);
+int acpi_power_resource_register_device(struct device *dev, acpi_handle handle);
+void acpi_power_resource_unregister_device(struct device *dev, acpi_handle handle);
#ifdef CONFIG_ACPI_PROC_EVENT
int acpi_bus_generate_proc_event(struct acpi_device *device, u8 type, int data);
int acpi_bus_generate_proc_event4(const char *class, const char *bid, u8 type, int data);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 46/76] ACPI: Introduce ACPI D3_COLD state support
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Len Brown, linux-kernel, Lin Ming
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Zhang Rui <rui.zhang@intel.com>
If a device has _PR3, it means the device supports D3_COLD.
Add the ability to validate and enter D3_COLD state in ACPI.
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/power.c | 4 ++--
drivers/acpi/scan.c | 7 +++++++
2 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/acpi/power.c b/drivers/acpi/power.c
index 9ac2a9f..0d681fb 100644
--- a/drivers/acpi/power.c
+++ b/drivers/acpi/power.c
@@ -500,14 +500,14 @@ int acpi_power_transition(struct acpi_device *device, int state)
{
int result;
- if (!device || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3))
+ if (!device || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3_COLD))
return -EINVAL;
if (device->power.state == state)
return 0;
if ((device->power.state < ACPI_STATE_D0)
- || (device->power.state > ACPI_STATE_D3))
+ || (device->power.state > ACPI_STATE_D3_COLD))
return -ENODEV;
/* TBD: Resources must be ordered. */
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 8ab80ba..571396c 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -885,6 +885,13 @@ static int acpi_bus_get_power_flags(struct acpi_device *device)
acpi_bus_add_power_resource(ps->resources.handles[j]);
}
+ /* The exist of _PR3 indicates D3Cold support */
+ if (i == ACPI_STATE_D3) {
+ status = acpi_get_handle(device->handle, object_name, &handle);
+ if (ACPI_SUCCESS(status))
+ device->power.states[ACPI_STATE_D3_COLD].flags.valid = 1;
+ }
+
/* Evaluate "_PSx" to see if we can do explicit sets */
object_name[2] = 'S';
status = acpi_get_handle(device->handle, object_name, &handle);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 45/76] ACPICA: Update to version 20120320
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Bob Moore, Lin Ming, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Bob Moore <robert.moore@intel.com>
Version 20120320.
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
include/acpi/acpixf.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/acpi/acpixf.h b/include/acpi/acpixf.h
index 488c986..9821101 100644
--- a/include/acpi/acpixf.h
+++ b/include/acpi/acpixf.h
@@ -47,7 +47,7 @@
/* Current ACPICA subsystem version in YYYYMMDD format */
-#define ACPI_CA_VERSION 0x20120215
+#define ACPI_CA_VERSION 0x20120320
#include "acconfig.h"
#include "actypes.h"
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 44/76] ACPICA: Object repair code: Support to add Package wrappers
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Len Brown, linux-kernel, Lin Ming, Bob Moore
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Bob Moore <robert.moore@intel.com>
Repair a common problem with objects that are defined to return
a variable-length Package of sub-objects. If there is only one
sub-object, some BIOS code mistakenly simply declares the single
object instead of a Package with one sub-object. This function
attempts to repair this error by wrapping a Package object around
the original object, creating the correct and expected Package
with one sub-object.
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/acpica/aclocal.h | 1 +
drivers/acpi/acpica/acnamesp.h | 5 +-
drivers/acpi/acpica/nspredef.c | 4 +-
drivers/acpi/acpica/nsrepair.c | 159 +++++++++++++++-------------------------
4 files changed, 64 insertions(+), 105 deletions(-)
diff --git a/drivers/acpi/acpica/aclocal.h b/drivers/acpi/acpica/aclocal.h
index 3f24068..e3922ca 100644
--- a/drivers/acpi/acpica/aclocal.h
+++ b/drivers/acpi/acpica/aclocal.h
@@ -370,6 +370,7 @@ struct acpi_predefined_data {
/* Defines for Flags field above */
#define ACPI_OBJECT_REPAIRED 1
+#define ACPI_OBJECT_WRAPPED 2
/*
* Bitmapped return value types
diff --git a/drivers/acpi/acpica/acnamesp.h b/drivers/acpi/acpica/acnamesp.h
index 2c9e0f0..9b19d4b 100644
--- a/drivers/acpi/acpica/acnamesp.h
+++ b/drivers/acpi/acpica/acnamesp.h
@@ -283,8 +283,9 @@ acpi_ns_repair_object(struct acpi_predefined_data *data,
union acpi_operand_object **return_object_ptr);
acpi_status
-acpi_ns_repair_package_list(struct acpi_predefined_data *data,
- union acpi_operand_object **obj_desc_ptr);
+acpi_ns_wrap_with_package(struct acpi_predefined_data *data,
+ union acpi_operand_object *original_object,
+ union acpi_operand_object **obj_desc_ptr);
acpi_status
acpi_ns_repair_null_element(struct acpi_predefined_data *data,
diff --git a/drivers/acpi/acpica/nspredef.c b/drivers/acpi/acpica/nspredef.c
index bbe46a4..23ce096 100644
--- a/drivers/acpi/acpica/nspredef.c
+++ b/drivers/acpi/acpica/nspredef.c
@@ -638,8 +638,8 @@ acpi_ns_check_package(struct acpi_predefined_data *data,
/* Create the new outer package and populate it */
status =
- acpi_ns_repair_package_list(data,
- return_object_ptr);
+ acpi_ns_wrap_with_package(data, *elements,
+ return_object_ptr);
if (ACPI_FAILURE(status)) {
return (status);
}
diff --git a/drivers/acpi/acpica/nsrepair.c b/drivers/acpi/acpica/nsrepair.c
index 9c35d20..5519a64 100644
--- a/drivers/acpi/acpica/nsrepair.c
+++ b/drivers/acpi/acpica/nsrepair.c
@@ -71,11 +71,10 @@ ACPI_MODULE_NAME("nsrepair")
* Buffer -> String
* Buffer -> Package of Integers
* Package -> Package of one Package
+ * An incorrect standalone object is wrapped with required outer package
*
* Additional possible repairs:
- *
* Required package elements that are NULL replaced by Integer/String/Buffer
- * Incorrect standalone package wrapped with required outer package
*
******************************************************************************/
/* Local prototypes */
@@ -91,10 +90,6 @@ static acpi_status
acpi_ns_convert_to_buffer(union acpi_operand_object *original_object,
union acpi_operand_object **return_object);
-static acpi_status
-acpi_ns_convert_to_package(union acpi_operand_object *original_object,
- union acpi_operand_object **return_object);
-
/*******************************************************************************
*
* FUNCTION: acpi_ns_repair_object
@@ -151,9 +146,24 @@ acpi_ns_repair_object(struct acpi_predefined_data *data,
}
}
if (expected_btypes & ACPI_RTYPE_PACKAGE) {
- status = acpi_ns_convert_to_package(return_object, &new_object);
+ /*
+ * A package is expected. We will wrap the existing object with a
+ * new package object. It is often the case that if a variable-length
+ * package is required, but there is only a single object needed, the
+ * BIOS will return that object instead of wrapping it with a Package
+ * object. Note: after the wrapping, the package will be validated
+ * for correct contents (expected object type or types).
+ */
+ status =
+ acpi_ns_wrap_with_package(data, return_object, &new_object);
if (ACPI_SUCCESS(status)) {
- goto object_repaired;
+ /*
+ * The original object just had its reference count
+ * incremented for being inserted into the new package.
+ */
+ *return_object_ptr = new_object; /* New Package object */
+ data->flags |= ACPI_OBJECT_REPAIRED;
+ return (AE_OK);
}
}
@@ -165,22 +175,27 @@ acpi_ns_repair_object(struct acpi_predefined_data *data,
/* Object was successfully repaired */
- /*
- * If the original object is a package element, we need to:
- * 1. Set the reference count of the new object to match the
- * reference count of the old object.
- * 2. Decrement the reference count of the original object.
- */
if (package_index != ACPI_NOT_PACKAGE_ELEMENT) {
- new_object->common.reference_count =
- return_object->common.reference_count;
+ /*
+ * The original object is a package element. We need to
+ * decrement the reference count of the original object,
+ * for removing it from the package.
+ *
+ * However, if the original object was just wrapped with a
+ * package object as part of the repair, we don't need to
+ * change the reference count.
+ */
+ if (!(data->flags & ACPI_OBJECT_WRAPPED)) {
+ new_object->common.reference_count =
+ return_object->common.reference_count;
- if (return_object->common.reference_count > 1) {
- return_object->common.reference_count--;
+ if (return_object->common.reference_count > 1) {
+ return_object->common.reference_count--;
+ }
}
ACPI_DEBUG_PRINT((ACPI_DB_REPAIR,
- "%s: Converted %s to expected %s at index %u\n",
+ "%s: Converted %s to expected %s at Package index %u\n",
data->pathname,
acpi_ut_get_object_type_name(return_object),
acpi_ut_get_object_type_name(new_object),
@@ -453,65 +468,6 @@ acpi_ns_convert_to_buffer(union acpi_operand_object *original_object,
/*******************************************************************************
*
- * FUNCTION: acpi_ns_convert_to_package
- *
- * PARAMETERS: original_object - Object to be converted
- * return_object - Where the new converted object is returned
- *
- * RETURN: Status. AE_OK if conversion was successful.
- *
- * DESCRIPTION: Attempt to convert a Buffer object to a Package. Each byte of
- * the buffer is converted to a single integer package element.
- *
- ******************************************************************************/
-
-static acpi_status
-acpi_ns_convert_to_package(union acpi_operand_object *original_object,
- union acpi_operand_object **return_object)
-{
- union acpi_operand_object *new_object;
- union acpi_operand_object **elements;
- u32 length;
- u8 *buffer;
-
- switch (original_object->common.type) {
- case ACPI_TYPE_BUFFER:
-
- /* Buffer-to-Package conversion */
-
- length = original_object->buffer.length;
- new_object = acpi_ut_create_package_object(length);
- if (!new_object) {
- return (AE_NO_MEMORY);
- }
-
- /* Convert each buffer byte to an integer package element */
-
- elements = new_object->package.elements;
- buffer = original_object->buffer.pointer;
-
- while (length--) {
- *elements =
- acpi_ut_create_integer_object((u64) *buffer);
- if (!*elements) {
- acpi_ut_remove_reference(new_object);
- return (AE_NO_MEMORY);
- }
- elements++;
- buffer++;
- }
- break;
-
- default:
- return (AE_AML_OPERAND_TYPE);
- }
-
- *return_object = new_object;
- return (AE_OK);
-}
-
-/*******************************************************************************
- *
* FUNCTION: acpi_ns_repair_null_element
*
* PARAMETERS: Data - Pointer to validation data structure
@@ -677,55 +633,56 @@ acpi_ns_remove_null_elements(struct acpi_predefined_data *data,
/*******************************************************************************
*
- * FUNCTION: acpi_ns_repair_package_list
+ * FUNCTION: acpi_ns_wrap_with_package
*
* PARAMETERS: Data - Pointer to validation data structure
- * obj_desc_ptr - Pointer to the object to repair. The new
- * package object is returned here,
- * overwriting the old object.
+ * original_object - Pointer to the object to repair.
+ * obj_desc_ptr - The new package object is returned here
*
* RETURN: Status, new object in *obj_desc_ptr
*
- * DESCRIPTION: Repair a common problem with objects that are defined to return
- * a variable-length Package of Packages. If the variable-length
- * is one, some BIOS code mistakenly simply declares a single
- * Package instead of a Package with one sub-Package. This
- * function attempts to repair this error by wrapping a Package
- * object around the original Package, creating the correct
- * Package with one sub-Package.
+ * DESCRIPTION: Repair a common problem with objects that are defined to
+ * return a variable-length Package of sub-objects. If there is
+ * only one sub-object, some BIOS code mistakenly simply declares
+ * the single object instead of a Package with one sub-object.
+ * This function attempts to repair this error by wrapping a
+ * Package object around the original object, creating the
+ * correct and expected Package with one sub-object.
*
* Names that can be repaired in this manner include:
- * _ALR, _CSD, _HPX, _MLS, _PRT, _PSS, _TRT, TSS
+ * _ALR, _CSD, _HPX, _MLS, _PLD, _PRT, _PSS, _TRT, _TSS,
+ * _BCL, _DOD, _FIX, _Sx
*
******************************************************************************/
acpi_status
-acpi_ns_repair_package_list(struct acpi_predefined_data *data,
- union acpi_operand_object **obj_desc_ptr)
+acpi_ns_wrap_with_package(struct acpi_predefined_data *data,
+ union acpi_operand_object *original_object,
+ union acpi_operand_object **obj_desc_ptr)
{
union acpi_operand_object *pkg_obj_desc;
- ACPI_FUNCTION_NAME(ns_repair_package_list);
+ ACPI_FUNCTION_NAME(ns_wrap_with_package);
/*
* Create the new outer package and populate it. The new package will
- * have a single element, the lone subpackage.
+ * have a single element, the lone sub-object.
*/
pkg_obj_desc = acpi_ut_create_package_object(1);
if (!pkg_obj_desc) {
return (AE_NO_MEMORY);
}
- pkg_obj_desc->package.elements[0] = *obj_desc_ptr;
+ pkg_obj_desc->package.elements[0] = original_object;
+
+ ACPI_DEBUG_PRINT((ACPI_DB_REPAIR,
+ "%s: Wrapped %s with expected Package object\n",
+ data->pathname,
+ acpi_ut_get_object_type_name(original_object)));
/* Return the new object in the object pointer */
*obj_desc_ptr = pkg_obj_desc;
- data->flags |= ACPI_OBJECT_REPAIRED;
-
- ACPI_DEBUG_PRINT((ACPI_DB_REPAIR,
- "%s: Repaired incorrectly formed Package\n",
- data->pathname));
-
+ data->flags |= ACPI_OBJECT_REPAIRED | ACPI_OBJECT_WRAPPED;
return (AE_OK);
}
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 43/76] ACPI: Make ACPI interrupt threaded
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Andi Kleen, Andi Kleen, linux-kernel, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Andi Kleen <andi@firstfloor.org>
Some ACPI interrupt actions may need to wait, and it's easiest to
have a thread context for this. So turn the ACPI interrupt
into a threaded interrupt.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/osl.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index 412a1e0..02367a8 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -595,7 +595,8 @@ acpi_os_install_interrupt_handler(u32 gsi, acpi_osd_handler handler,
acpi_irq_handler = handler;
acpi_irq_context = context;
- if (request_irq(irq, acpi_irq, IRQF_SHARED, "acpi", acpi_irq)) {
+ if (request_threaded_irq(irq, NULL, acpi_irq, IRQF_SHARED, "acpi",
+ acpi_irq)) {
printk(KERN_ERR PREFIX "SCI (IRQ%d) allocation failed\n", irq);
acpi_irq_handler = NULL;
return AE_NOT_ACQUIRED;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 42/76] ACPI: ec: Do request_region outside WARN()
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Andi Kleen, Andi Kleen, linux-kernel, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Andi Kleen <andi@firstfloor.org>
WARN() is not supposed to have side effects, so move the request_regions
outside.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/ec.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
index b19a18d..3268dcf 100644
--- a/drivers/acpi/ec.c
+++ b/drivers/acpi/ec.c
@@ -812,10 +812,10 @@ static int acpi_ec_add(struct acpi_device *device)
first_ec = ec;
device->driver_data = ec;
- WARN(!request_region(ec->data_addr, 1, "EC data"),
- "Could not request EC data io port 0x%lx", ec->data_addr);
- WARN(!request_region(ec->command_addr, 1, "EC cmd"),
- "Could not request EC cmd io port 0x%lx", ec->command_addr);
+ ret = !!request_region(ec->data_addr, 1, "EC data");
+ WARN(!ret, "Could not request EC data io port 0x%lx", ec->data_addr);
+ ret = !!request_region(ec->command_addr, 1, "EC cmd");
+ WARN(!ret, "Could not request EC cmd io port 0x%lx", ec->command_addr);
pr_info(PREFIX "GPE = 0x%lx, I/O: command/status = 0x%lx, data = 0x%lx\n",
ec->gpe, ec->command_addr, ec->data_addr);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 41/76] tools turbostat: harden against cpu online/offline
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Len Brown <len.brown@intel.com>
Sometimes users have turbostat running in interval mode
when they take processors offline/online.
Previously, turbostat would survive, but not gracefully.
Tighten up the error checking so turbostat notices
changesn sooner, and print just 1 line on change:
turbostat: re-initialized with num_cpus %d
Signed-off-by: Len Brown <len.brown@intel.com>
---
tools/power/x86/turbostat/turbostat.c | 115 +++++++++++++++++----------------
1 file changed, 61 insertions(+), 54 deletions(-)
diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c
index fa60872..ab2f682 100644
--- a/tools/power/x86/turbostat/turbostat.c
+++ b/tools/power/x86/turbostat/turbostat.c
@@ -71,7 +71,6 @@ unsigned int show_cpu;
int aperf_mperf_unstable;
int backwards_count;
char *progname;
-int need_reinitialize;
int num_cpus;
cpu_set_t *cpu_mask;
@@ -138,30 +137,24 @@ int cpu_migrate(int cpu)
return 0;
}
-unsigned long long get_msr(int cpu, off_t offset)
+int get_msr(int cpu, off_t offset, unsigned long long *msr)
{
ssize_t retval;
- unsigned long long msr;
char pathname[32];
int fd;
sprintf(pathname, "/dev/cpu/%d/msr", cpu);
fd = open(pathname, O_RDONLY);
- if (fd < 0) {
- perror(pathname);
- need_reinitialize = 1;
- return 0;
- }
-
- retval = pread(fd, &msr, sizeof msr, offset);
- if (retval != sizeof msr) {
- fprintf(stderr, "cpu%d pread(..., 0x%zx) = %jd\n",
- cpu, offset, retval);
- exit(-2);
- }
+ if (fd < 0)
+ return -1;
+ retval = pread(fd, msr, sizeof *msr, offset);
close(fd);
- return msr;
+
+ if (retval != sizeof *msr)
+ return -1;
+
+ return 0;
}
void print_header(void)
@@ -506,36 +499,51 @@ void compute_average(struct counters *delta, struct counters *avg)
free(sum);
}
-void get_counters(struct counters *cnt)
+int get_counters(struct counters *cnt)
{
for ( ; cnt; cnt = cnt->next) {
- if (cpu_migrate(cnt->cpu)) {
- need_reinitialize = 1;
- return;
+
+ if (cpu_migrate(cnt->cpu))
+ return -1;
+
+ if (get_msr(cnt->cpu, MSR_TSC, &cnt->tsc))
+ return -1;
+
+ if (has_aperf) {
+ if (get_msr(cnt->cpu, MSR_APERF, &cnt->aperf))
+ return -1;
+ if (get_msr(cnt->cpu, MSR_MPERF, &cnt->mperf))
+ return -1;
+ }
+
+ if (do_nhm_cstates) {
+ if (get_msr(cnt->cpu, MSR_CORE_C3_RESIDENCY, &cnt->c3))
+ return -1;
+ if (get_msr(cnt->cpu, MSR_CORE_C6_RESIDENCY, &cnt->c6))
+ return -1;
}
- cnt->tsc = get_msr(cnt->cpu, MSR_TSC);
- if (do_nhm_cstates)
- cnt->c3 = get_msr(cnt->cpu, MSR_CORE_C3_RESIDENCY);
- if (do_nhm_cstates)
- cnt->c6 = get_msr(cnt->cpu, MSR_CORE_C6_RESIDENCY);
- if (do_snb_cstates)
- cnt->c7 = get_msr(cnt->cpu, MSR_CORE_C7_RESIDENCY);
- if (has_aperf)
- cnt->aperf = get_msr(cnt->cpu, MSR_APERF);
- if (has_aperf)
- cnt->mperf = get_msr(cnt->cpu, MSR_MPERF);
- if (do_snb_cstates)
- cnt->pc2 = get_msr(cnt->cpu, MSR_PKG_C2_RESIDENCY);
- if (do_nhm_cstates)
- cnt->pc3 = get_msr(cnt->cpu, MSR_PKG_C3_RESIDENCY);
- if (do_nhm_cstates)
- cnt->pc6 = get_msr(cnt->cpu, MSR_PKG_C6_RESIDENCY);
if (do_snb_cstates)
- cnt->pc7 = get_msr(cnt->cpu, MSR_PKG_C7_RESIDENCY);
+ if (get_msr(cnt->cpu, MSR_CORE_C7_RESIDENCY, &cnt->c7))
+ return -1;
+
+ if (do_nhm_cstates) {
+ if (get_msr(cnt->cpu, MSR_PKG_C3_RESIDENCY, &cnt->pc3))
+ return -1;
+ if (get_msr(cnt->cpu, MSR_PKG_C6_RESIDENCY, &cnt->pc6))
+ return -1;
+ }
+ if (do_snb_cstates) {
+ if (get_msr(cnt->cpu, MSR_PKG_C2_RESIDENCY, &cnt->pc2))
+ return -1;
+ if (get_msr(cnt->cpu, MSR_PKG_C7_RESIDENCY, &cnt->pc7))
+ return -1;
+ }
if (extra_msr_offset)
- cnt->extra_msr = get_msr(cnt->cpu, extra_msr_offset);
+ if (get_msr(cnt->cpu, extra_msr_offset, &cnt->extra_msr))
+ return -1;
}
+ return 0;
}
void print_nehalem_info(void)
@@ -546,7 +554,7 @@ void print_nehalem_info(void)
if (!do_nehalem_platform_info)
return;
- msr = get_msr(0, MSR_NEHALEM_PLATFORM_INFO);
+ get_msr(0, MSR_NEHALEM_PLATFORM_INFO, &msr);
ratio = (msr >> 40) & 0xFF;
fprintf(stderr, "%d * %.0f = %.0f MHz max efficiency\n",
@@ -562,7 +570,7 @@ void print_nehalem_info(void)
if (!do_nehalem_turbo_ratio_limit)
return;
- msr = get_msr(0, MSR_NEHALEM_TURBO_RATIO_LIMIT);
+ get_msr(0, MSR_NEHALEM_TURBO_RATIO_LIMIT, &msr);
ratio = (msr >> 24) & 0xFF;
if (ratio)
@@ -755,7 +763,7 @@ int get_core_id(int cpu)
}
/*
- * run func(index, cpu) on every cpu in /proc/stat
+ * run func(pkg, core, cpu) on every cpu in /proc/stat
*/
int for_all_cpus(void (func)(int, int, int))
@@ -791,20 +799,18 @@ int for_all_cpus(void (func)(int, int, int))
void re_initialize(void)
{
- printf("turbostat: topology changed, re-initializing.\n");
free_all_counters();
num_cpus = for_all_cpus(alloc_new_counters);
- need_reinitialize = 0;
cpu_mask_uninit();
cpu_mask_init(num_cpus);
- printf("num_cpus is now %d\n", num_cpus);
+ printf("turbostat: re-initialized with num_cpus %d\n", num_cpus);
}
void dummy(int pkg, int core, int cpu) { return; }
/*
* check to see if a cpu came on-line
*/
-void verify_num_cpus(void)
+int verify_num_cpus(void)
{
int new_num_cpus;
@@ -814,8 +820,9 @@ void verify_num_cpus(void)
if (verbose)
printf("num_cpus was %d, is now %d\n",
num_cpus, new_num_cpus);
- need_reinitialize = 1;
+ return -1;
}
+ return 0;
}
void turbostat_loop()
@@ -825,25 +832,25 @@ restart:
gettimeofday(&tv_even, (struct timezone *)NULL);
while (1) {
- verify_num_cpus();
- if (need_reinitialize) {
+ if (verify_num_cpus()) {
re_initialize();
goto restart;
}
sleep(interval_sec);
- get_counters(cnt_odd);
+ if (get_counters(cnt_odd)) {
+ re_initialize();
+ goto restart;
+ }
gettimeofday(&tv_odd, (struct timezone *)NULL);
-
compute_delta(cnt_odd, cnt_even, cnt_delta);
timersub(&tv_odd, &tv_even, &tv_delta);
compute_average(cnt_delta, cnt_average);
print_counters(cnt_delta);
- if (need_reinitialize) {
+ sleep(interval_sec);
+ if (get_counters(cnt_even)) {
re_initialize();
goto restart;
}
- sleep(interval_sec);
- get_counters(cnt_even);
gettimeofday(&tv_even, (struct timezone *)NULL);
compute_delta(cnt_even, cnt_odd, cnt_delta);
timersub(&tv_even, &tv_odd, &tv_delta);
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 40/76] tools turbostat: reduce measurement overhead due to IPIs
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Len Brown, Youquan Song
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Len Brown <len.brown@intel.com>
turbostat uses /dev/cpu/*/msr interface to read MSRs.
For modern systems, it reads 10 MSR/CPU. This can
be observed as 10 "Function Call Interrupts"
per CPU per sample added to /proc/interrupts.
This overhead is measurable on large idle systems,
and as Yoquan Song pointed out, it can even trick
cpuidle into thinking the system is busy.
Here turbostat re-schedules itself in-turn to each
CPU so that its MSR reads will always be local.
This replaces the 10 "Function Call Interrupts"
with a single "Rescheduling interrupt" per sample
per CPU.
On an idle 32-CPU system, this shifts some residency from
the shallow c1 state to the deeper c7 state:
# ./turbostat.old -s
%c0 GHz TSC %c1 %c3 %c6 %c7 %pc2 %pc3 %pc6 %pc7
0.27 1.29 2.29 0.95 0.02 0.00 98.77 20.23 0.00 77.41 0.00
0.25 1.24 2.29 0.98 0.02 0.00 98.75 20.34 0.03 77.74 0.00
0.27 1.22 2.29 0.54 0.00 0.00 99.18 20.64 0.00 77.70 0.00
0.26 1.22 2.29 1.22 0.00 0.00 98.52 20.22 0.00 77.74 0.00
0.26 1.38 2.29 0.78 0.02 0.00 98.95 20.51 0.05 77.56 0.00
^C
i# ./turbostat.new -s
%c0 GHz TSC %c1 %c3 %c6 %c7 %pc2 %pc3 %pc6 %pc7
0.27 1.20 2.29 0.24 0.01 0.00 99.49 20.58 0.00 78.20 0.00
0.27 1.22 2.29 0.25 0.00 0.00 99.48 20.79 0.00 77.85 0.00
0.27 1.20 2.29 0.25 0.02 0.00 99.46 20.71 0.03 77.89 0.00
0.28 1.26 2.29 0.25 0.01 0.00 99.46 20.89 0.02 77.67 0.00
0.27 1.20 2.29 0.24 0.01 0.00 99.48 20.65 0.00 78.04 0.00
cc: Youquan Song <youquan.song@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
tools/power/x86/turbostat/turbostat.c | 46 +++++++++++++++++++++++++++++++++
1 file changed, 46 insertions(+)
diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c
index 6436d54..fa60872 100644
--- a/tools/power/x86/turbostat/turbostat.c
+++ b/tools/power/x86/turbostat/turbostat.c
@@ -19,6 +19,7 @@
* 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
*/
+#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <sys/types.h>
@@ -32,6 +33,7 @@
#include <dirent.h>
#include <string.h>
#include <ctype.h>
+#include <sched.h>
#define MSR_TSC 0x10
#define MSR_NEHALEM_PLATFORM_INFO 0xCE
@@ -72,6 +74,8 @@ char *progname;
int need_reinitialize;
int num_cpus;
+cpu_set_t *cpu_mask;
+size_t cpu_mask_size;
struct counters {
unsigned long long tsc; /* per thread */
@@ -100,6 +104,40 @@ struct timeval tv_even;
struct timeval tv_odd;
struct timeval tv_delta;
+/*
+ * cpu_mask_init(ncpus)
+ *
+ * allocate and clear cpu_mask
+ * set cpu_mask_size
+ */
+void cpu_mask_init(int ncpus)
+{
+ cpu_mask = CPU_ALLOC(ncpus);
+ if (cpu_mask == NULL) {
+ perror("CPU_ALLOC");
+ exit(3);
+ }
+ cpu_mask_size = CPU_ALLOC_SIZE(ncpus);
+ CPU_ZERO_S(cpu_mask_size, cpu_mask);
+}
+
+void cpu_mask_uninit()
+{
+ CPU_FREE(cpu_mask);
+ cpu_mask = NULL;
+ cpu_mask_size = 0;
+}
+
+int cpu_migrate(int cpu)
+{
+ CPU_ZERO_S(cpu_mask_size, cpu_mask);
+ CPU_SET_S(cpu, cpu_mask_size, cpu_mask);
+ if (sched_setaffinity(0, cpu_mask_size, cpu_mask) == -1)
+ return -1;
+ else
+ return 0;
+}
+
unsigned long long get_msr(int cpu, off_t offset)
{
ssize_t retval;
@@ -471,6 +509,11 @@ void compute_average(struct counters *delta, struct counters *avg)
void get_counters(struct counters *cnt)
{
for ( ; cnt; cnt = cnt->next) {
+ if (cpu_migrate(cnt->cpu)) {
+ need_reinitialize = 1;
+ return;
+ }
+
cnt->tsc = get_msr(cnt->cpu, MSR_TSC);
if (do_nhm_cstates)
cnt->c3 = get_msr(cnt->cpu, MSR_CORE_C3_RESIDENCY);
@@ -752,6 +795,8 @@ void re_initialize(void)
free_all_counters();
num_cpus = for_all_cpus(alloc_new_counters);
need_reinitialize = 0;
+ cpu_mask_uninit();
+ cpu_mask_init(num_cpus);
printf("num_cpus is now %d\n", num_cpus);
}
@@ -984,6 +1029,7 @@ void turbostat_init()
check_super_user();
num_cpus = for_all_cpus(alloc_new_counters);
+ cpu_mask_init(num_cpus);
if (verbose)
print_nehalem_info();
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 39/76] tools turbostat: add summary option
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: linux-kernel, Len Brown
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Len Brown <len.brown@intel.com>
turbostat -s
cuts down on the amount of output, per user request.
also treak some output whitespace and the man page.
Signed-off-by: Len Brown <len.brown@intel.com>
---
tools/power/x86/turbostat/turbostat.8 | 99 +++++++++++++++++++--------------
tools/power/x86/turbostat/turbostat.c | 90 ++++++++++++++++++++----------
2 files changed, 120 insertions(+), 69 deletions(-)
diff --git a/tools/power/x86/turbostat/turbostat.8 b/tools/power/x86/turbostat/turbostat.8
index 555c69a..adf175f 100644
--- a/tools/power/x86/turbostat/turbostat.8
+++ b/tools/power/x86/turbostat/turbostat.8
@@ -4,11 +4,13 @@ turbostat \- Report processor frequency and idle statistics
.SH SYNOPSIS
.ft B
.B turbostat
+.RB [ "\-s" ]
.RB [ "\-v" ]
.RB [ "\-M MSR#" ]
.RB command
.br
.B turbostat
+.RB [ "\-s" ]
.RB [ "\-v" ]
.RB [ "\-M MSR#" ]
.RB [ "\-i interval_sec" ]
@@ -25,6 +27,8 @@ supports an "invariant" TSC, plus the APERF and MPERF MSRs.
on processors that additionally support C-state residency counters.
.SS Options
+The \fB-s\fP option prints only a 1-line summary for each sample interval.
+.PP
The \fB-v\fP option increases verbosity.
.PP
The \fB-M MSR#\fP option dumps the specified MSR,
@@ -39,13 +43,14 @@ displays the statistics gathered since it was forked.
.SH FIELD DESCRIPTIONS
.nf
\fBpk\fP processor package number.
-\fBcr\fP processor core number.
+\fBcor\fP processor core number.
\fBCPU\fP Linux CPU (logical processor) number.
+Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading Technology.
\fB%c0\fP percent of the interval that the CPU retired instructions.
\fBGHz\fP average clock rate while the CPU was in c0 state.
\fBTSC\fP average GHz that the TSC ran during the entire interval.
-\fB%c1, %c3, %c6\fP show the percentage residency in hardware core idle states.
-\fB%pc3, %pc6\fP percentage residency in hardware package idle states.
+\fB%c1, %c3, %c6, %c7\fP show the percentage residency in hardware core idle states.
+\fB%pc2, %pc3, %pc6, %pc7\fP percentage residency in hardware package idle states.
.fi
.PP
.SH EXAMPLE
@@ -53,25 +58,37 @@ Without any parameters, turbostat prints out counters ever 5 seconds.
(override interval with "-i sec" option, or specify a command
for turbostat to fork).
-The first row of statistics reflect the average for the entire system.
+The first row of statistics is a summary for the entire system.
+Note that the summary is a weighted average.
Subsequent rows show per-CPU statistics.
.nf
[root@x980]# ./turbostat
-cr CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
- 0.04 1.62 3.38 0.11 0.00 99.85 0.00 95.07
- 0 0 0.04 1.62 3.38 0.06 0.00 99.90 0.00 95.07
- 0 6 0.02 1.62 3.38 0.08 0.00 99.90 0.00 95.07
- 1 2 0.10 1.62 3.38 0.29 0.00 99.61 0.00 95.07
- 1 8 0.11 1.62 3.38 0.28 0.00 99.61 0.00 95.07
- 2 4 0.01 1.62 3.38 0.01 0.00 99.98 0.00 95.07
- 2 10 0.01 1.61 3.38 0.02 0.00 99.98 0.00 95.07
- 8 1 0.07 1.62 3.38 0.15 0.00 99.78 0.00 95.07
- 8 7 0.03 1.62 3.38 0.19 0.00 99.78 0.00 95.07
- 9 3 0.01 1.62 3.38 0.02 0.00 99.98 0.00 95.07
- 9 9 0.01 1.62 3.38 0.02 0.00 99.98 0.00 95.07
- 10 5 0.01 1.62 3.38 0.13 0.00 99.86 0.00 95.07
- 10 11 0.08 1.62 3.38 0.05 0.00 99.86 0.00 95.07
+cor CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
+ 0.60 1.63 3.38 2.91 0.00 96.49 0.00 76.64
+ 0 0 0.59 1.62 3.38 4.51 0.00 94.90 0.00 76.64
+ 0 6 1.13 1.64 3.38 3.97 0.00 94.90 0.00 76.64
+ 1 2 0.08 1.62 3.38 0.07 0.00 99.85 0.00 76.64
+ 1 8 0.03 1.62 3.38 0.12 0.00 99.85 0.00 76.64
+ 2 4 0.01 1.62 3.38 0.06 0.00 99.93 0.00 76.64
+ 2 10 0.04 1.62 3.38 0.02 0.00 99.93 0.00 76.64
+ 8 1 2.85 1.62 3.38 11.71 0.00 85.44 0.00 76.64
+ 8 7 1.98 1.62 3.38 12.58 0.00 85.44 0.00 76.64
+ 9 3 0.36 1.62 3.38 0.71 0.00 98.93 0.00 76.64
+ 9 9 0.09 1.62 3.38 0.98 0.00 98.93 0.00 76.64
+ 10 5 0.03 1.62 3.38 0.09 0.00 99.87 0.00 76.64
+ 10 11 0.07 1.62 3.38 0.06 0.00 99.87 0.00 76.64
+.fi
+.SH SUMMARY EXAMPLE
+The "-s" option prints the column headers just once,
+and then the one line system summary for each sample interval.
+
+.nf
+[root@x980]# ./turbostat -s
+ %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
+ 0.61 1.89 3.38 5.95 0.00 93.44 0.00 66.33
+ 0.52 1.62 3.38 6.83 0.00 92.65 0.00 61.11
+ 0.62 1.92 3.38 5.47 0.00 93.91 0.00 67.31
.fi
.SH VERBOSE EXAMPLE
The "-v" option adds verbosity to the output:
@@ -101,33 +118,33 @@ until ^C while the other CPUs are mostly idle:
.nf
[root@x980 lenb]# ./turbostat cat /dev/zero > /dev/null
-
-^Ccr CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
- 8.49 3.63 3.38 16.23 0.66 74.63 0.00 0.00
- 0 0 1.22 3.62 3.38 32.18 0.00 66.60 0.00 0.00
- 0 6 0.40 3.61 3.38 33.00 0.00 66.60 0.00 0.00
- 1 2 0.11 3.14 3.38 0.19 3.95 95.75 0.00 0.00
- 1 8 0.05 2.88 3.38 0.25 3.95 95.75 0.00 0.00
- 2 4 0.00 3.13 3.38 0.02 0.00 99.98 0.00 0.00
- 2 10 0.00 3.09 3.38 0.02 0.00 99.98 0.00 0.00
- 8 1 0.04 3.50 3.38 14.43 0.00 85.54 0.00 0.00
- 8 7 0.03 2.98 3.38 14.43 0.00 85.54 0.00 0.00
- 9 3 0.00 3.16 3.38 100.00 0.00 0.00 0.00 0.00
- 9 9 99.93 3.63 3.38 0.06 0.00 0.00 0.00 0.00
- 10 5 0.01 2.82 3.38 0.08 0.00 99.91 0.00 0.00
- 10 11 0.02 3.36 3.38 0.06 0.00 99.91 0.00 0.00
-6.950866 sec
+^C
+cor CPU %c0 GHz TSC %c1 %c3 %c6 %pc3 %pc6
+ 8.63 3.64 3.38 14.46 0.49 76.42 0.00 0.00
+ 0 0 0.34 3.36 3.38 99.66 0.00 0.00 0.00 0.00
+ 0 6 99.96 3.64 3.38 0.04 0.00 0.00 0.00 0.00
+ 1 2 0.14 3.50 3.38 1.75 2.04 96.07 0.00 0.00
+ 1 8 0.38 3.57 3.38 1.51 2.04 96.07 0.00 0.00
+ 2 4 0.01 2.65 3.38 0.06 0.00 99.93 0.00 0.00
+ 2 10 0.03 2.12 3.38 0.04 0.00 99.93 0.00 0.00
+ 8 1 0.91 3.59 3.38 35.27 0.92 62.90 0.00 0.00
+ 8 7 1.61 3.63 3.38 34.57 0.92 62.90 0.00 0.00
+ 9 3 0.04 3.38 3.38 0.20 0.00 99.76 0.00 0.00
+ 9 9 0.04 3.29 3.38 0.20 0.00 99.76 0.00 0.00
+ 10 5 0.03 3.08 3.38 0.12 0.00 99.85 0.00 0.00
+ 10 11 0.05 3.07 3.38 0.10 0.00 99.85 0.00 0.00
+4.907015 sec
.fi
-Above the cycle soaker drives cpu9 up 3.6 Ghz turbo limit
+Above the cycle soaker drives cpu6 up 3.6 Ghz turbo limit
while the other processors are generally in various states of idle.
-Note that cpu3 is an HT sibling sharing core9
-with cpu9, and thus it is unable to get to an idle state
-deeper than c1 while cpu9 is busy.
+Note that cpu0 is an HT sibling sharing core0
+with cpu6, and thus it is unable to get to an idle state
+deeper than c1 while cpu6 is busy.
-Note that turbostat reports average GHz of 3.61, while
-the arithmetic average of the GHz column above is 3.24.
+Note that turbostat reports average GHz of 3.64, while
+the arithmetic average of the GHz column above is lower.
This is a weighted average, where the weight is %c0. ie. it is the total number of
un-halted cycles elapsed per time divided by the number of CPUs.
.SH NOTES
@@ -167,6 +184,6 @@ http://www.intel.com/products/processor/manuals/
.SH "SEE ALSO"
msr(4), vmstat(8)
.PP
-.SH AUTHORS
+.SH AUTHOR
.nf
Written by Len Brown <len.brown@intel.com>
diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c
index 310d3dd..6436d54 100644
--- a/tools/power/x86/turbostat/turbostat.c
+++ b/tools/power/x86/turbostat/turbostat.c
@@ -2,7 +2,7 @@
* turbostat -- show CPU frequency and C-state residency
* on modern Intel turbo-capable processors.
*
- * Copyright (c) 2010, Intel Corporation.
+ * Copyright (c) 2012 Intel Corporation.
* Len Brown <len.brown@intel.com>
*
* This program is free software; you can redistribute it and/or modify it
@@ -49,6 +49,7 @@
char *proc_stat = "/proc/stat";
unsigned int interval_sec = 5; /* set with -i interval_sec */
unsigned int verbose; /* set with -v */
+unsigned int summary_only; /* set with -s */
unsigned int skip_c0;
unsigned int skip_c1;
unsigned int do_nhm_cstates;
@@ -129,14 +130,18 @@ void print_header(void)
{
if (show_pkg)
fprintf(stderr, "pk");
+ if (show_pkg)
+ fprintf(stderr, " ");
if (show_core)
- fprintf(stderr, " cr");
+ fprintf(stderr, "cor");
if (show_cpu)
fprintf(stderr, " CPU");
+ if (show_pkg || show_core || show_cpu)
+ fprintf(stderr, " ");
if (do_nhm_cstates)
- fprintf(stderr, " %%c0 ");
+ fprintf(stderr, " %%c0");
if (has_aperf)
- fprintf(stderr, " GHz");
+ fprintf(stderr, " GHz");
fprintf(stderr, " TSC");
if (do_nhm_cstates)
fprintf(stderr, " %%c1");
@@ -147,13 +152,13 @@ void print_header(void)
if (do_snb_cstates)
fprintf(stderr, " %%c7");
if (do_snb_cstates)
- fprintf(stderr, " %%pc2");
+ fprintf(stderr, " %%pc2");
if (do_nhm_cstates)
- fprintf(stderr, " %%pc3");
+ fprintf(stderr, " %%pc3");
if (do_nhm_cstates)
- fprintf(stderr, " %%pc6");
+ fprintf(stderr, " %%pc6");
if (do_snb_cstates)
- fprintf(stderr, " %%pc7");
+ fprintf(stderr, " %%pc7");
if (extra_msr_offset)
fprintf(stderr, " MSR 0x%x ", extra_msr_offset);
@@ -187,6 +192,15 @@ void dump_list(struct counters *cnt)
dump_cnt(cnt);
}
+/*
+ * column formatting convention & formats
+ * package: "pk" 2 columns %2d
+ * core: "cor" 3 columns %3d
+ * CPU: "CPU" 3 columns %3d
+ * GHz: "GHz" 3 columns %3.2
+ * TSC: "TSC" 3 columns %3.2
+ * percentage " %pc3" %6.2
+ */
void print_cnt(struct counters *p)
{
double interval_float;
@@ -196,39 +210,45 @@ void print_cnt(struct counters *p)
/* topology columns, print blanks on 1st (average) line */
if (p == cnt_average) {
if (show_pkg)
+ fprintf(stderr, " ");
+ if (show_pkg && show_core)
fprintf(stderr, " ");
if (show_core)
- fprintf(stderr, " ");
+ fprintf(stderr, " ");
if (show_cpu)
- fprintf(stderr, " ");
+ fprintf(stderr, " " " ");
} else {
if (show_pkg)
- fprintf(stderr, "%d", p->pkg);
+ fprintf(stderr, "%2d", p->pkg);
+ if (show_pkg && show_core)
+ fprintf(stderr, " ");
if (show_core)
- fprintf(stderr, "%4d", p->core);
+ fprintf(stderr, "%3d", p->core);
if (show_cpu)
- fprintf(stderr, "%4d", p->cpu);
+ fprintf(stderr, " %3d", p->cpu);
}
/* %c0 */
if (do_nhm_cstates) {
+ if (show_pkg || show_core || show_cpu)
+ fprintf(stderr, " ");
if (!skip_c0)
- fprintf(stderr, "%7.2f", 100.0 * p->mperf/p->tsc);
+ fprintf(stderr, "%6.2f", 100.0 * p->mperf/p->tsc);
else
- fprintf(stderr, " ****");
+ fprintf(stderr, " ****");
}
/* GHz */
if (has_aperf) {
if (!aperf_mperf_unstable) {
- fprintf(stderr, "%5.2f",
+ fprintf(stderr, " %3.2f",
1.0 * p->tsc / units * p->aperf /
p->mperf / interval_float);
} else {
if (p->aperf > p->tsc || p->mperf > p->tsc) {
- fprintf(stderr, " ****");
+ fprintf(stderr, " ***");
} else {
- fprintf(stderr, "%4.1f*",
+ fprintf(stderr, "%3.1f*",
1.0 * p->tsc /
units * p->aperf /
p->mperf / interval_float);
@@ -241,7 +261,7 @@ void print_cnt(struct counters *p)
if (do_nhm_cstates) {
if (!skip_c1)
- fprintf(stderr, "%7.2f", 100.0 * p->c1/p->tsc);
+ fprintf(stderr, " %6.2f", 100.0 * p->c1/p->tsc);
else
fprintf(stderr, " ****");
}
@@ -252,13 +272,13 @@ void print_cnt(struct counters *p)
if (do_snb_cstates)
fprintf(stderr, " %6.2f", 100.0 * p->c7/p->tsc);
if (do_snb_cstates)
- fprintf(stderr, " %5.2f", 100.0 * p->pc2/p->tsc);
+ fprintf(stderr, " %6.2f", 100.0 * p->pc2/p->tsc);
if (do_nhm_cstates)
- fprintf(stderr, " %5.2f", 100.0 * p->pc3/p->tsc);
+ fprintf(stderr, " %6.2f", 100.0 * p->pc3/p->tsc);
if (do_nhm_cstates)
- fprintf(stderr, " %5.2f", 100.0 * p->pc6/p->tsc);
+ fprintf(stderr, " %6.2f", 100.0 * p->pc6/p->tsc);
if (do_snb_cstates)
- fprintf(stderr, " %5.2f", 100.0 * p->pc7/p->tsc);
+ fprintf(stderr, " %6.2f", 100.0 * p->pc7/p->tsc);
if (extra_msr_offset)
fprintf(stderr, " 0x%016llx", p->extra_msr);
putc('\n', stderr);
@@ -267,12 +287,20 @@ void print_cnt(struct counters *p)
void print_counters(struct counters *counters)
{
struct counters *cnt;
+ static int printed;
- print_header();
+
+ if (!printed || !summary_only)
+ print_header();
if (num_cpus > 1)
print_cnt(cnt_average);
+ printed = 1;
+
+ if (summary_only)
+ return;
+
for (cnt = counters; cnt != NULL; cnt = cnt->next)
print_cnt(cnt);
@@ -557,7 +585,8 @@ void insert_counters(struct counters **list,
return;
}
- show_cpu = 1; /* there is more than one CPU */
+ if (!summary_only)
+ show_cpu = 1; /* there is more than one CPU */
/*
* insert on front of list.
@@ -575,13 +604,15 @@ void insert_counters(struct counters **list,
while (prev->next && (prev->next->pkg < new->pkg)) {
prev = prev->next;
- show_pkg = 1; /* there is more than 1 package */
+ if (!summary_only)
+ show_pkg = 1; /* there is more than 1 package */
}
while (prev->next && (prev->next->pkg == new->pkg)
&& (prev->next->core < new->core)) {
prev = prev->next;
- show_core = 1; /* there is more than 1 core */
+ if (!summary_only)
+ show_core = 1; /* there is more than 1 core */
}
while (prev->next && (prev->next->pkg == new->pkg)
@@ -1005,8 +1036,11 @@ void cmdline(int argc, char **argv)
progname = argv[0];
- while ((opt = getopt(argc, argv, "+vi:M:")) != -1) {
+ while ((opt = getopt(argc, argv, "+svi:M:")) != -1) {
switch (opt) {
+ case 's':
+ summary_only++;
+ break;
case 'v':
verbose++;
break;
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
* [PATCH 38/76] ACPI: Move module parameter gts and bfs to sleep.c
From: Len Brown @ 2012-03-30 10:13 UTC (permalink / raw)
To: linux-acpi, linux-pm; +Cc: Len Brown, linux-kernel, Lin Ming
In-Reply-To: <09f98a825a821f7a3f1b162f9ed023f37213a63b.1333101989.git.len.brown@intel.com>
From: Lin Ming <ming.m.lin@intel.com>
Move linux specific module parameter gts and bfs out of ACPICA core
code to sleep.c.
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/acpi/acpica/hwesleep.c | 13 -------------
drivers/acpi/sleep.c | 38 +++++++++++++++++++++++++++++++-------
2 files changed, 31 insertions(+), 20 deletions(-)
diff --git a/drivers/acpi/acpica/hwesleep.c b/drivers/acpi/acpica/hwesleep.c
index 1c409e8..29e8592 100644
--- a/drivers/acpi/acpica/hwesleep.c
+++ b/drivers/acpi/acpica/hwesleep.c
@@ -44,17 +44,10 @@
#include <acpi/acpi.h>
#include "accommon.h"
-#include <linux/module.h>
#define _COMPONENT ACPI_HARDWARE
ACPI_MODULE_NAME("hwesleep")
-static unsigned int gts, bfs;
-module_param(gts, uint, 0644);
-module_param(bfs, uint, 0644);
-MODULE_PARM_DESC(gts, "Enable evaluation of _GTS on suspend.");
-MODULE_PARM_DESC(bfs, "Enable evaluation of _BFS on resume".);
-
/*******************************************************************************
*
* FUNCTION: acpi_hw_execute_sleep_method
@@ -76,12 +69,6 @@ void acpi_hw_execute_sleep_method(char *method_pathname, u32 integer_argument)
ACPI_FUNCTION_TRACE(hw_execute_sleep_method);
- if (!ACPI_STRCMP(METHOD_PATHNAME__GTS, method_pathname) && !gts)
- return_VOID;
-
- if (!ACPI_STRCMP(METHOD_PATHNAME__BFS, method_pathname) && !bfs)
- return_VOID;
-
/* One argument, integer_argument; No return value expected */
arg_list.count = 1;
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index 8f1fb45..acf81fe 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -17,6 +17,7 @@
#include <linux/suspend.h>
#include <linux/reboot.h>
#include <linux/acpi.h>
+#include <linux/module.h>
#include <asm/io.h>
@@ -26,6 +27,24 @@
#include "internal.h"
#include "sleep.h"
+static unsigned int gts, bfs;
+module_param(gts, uint, 0644);
+module_param(bfs, uint, 0644);
+MODULE_PARM_DESC(gts, "Enable evaluation of _GTS on suspend.");
+MODULE_PARM_DESC(bfs, "Enable evaluation of _BFS on resume".);
+
+static u8 wake_sleep_flags(void)
+{
+ u8 flags = ACPI_NO_OPTIONAL_METHODS;
+
+ if (gts)
+ flags |= ACPI_EXECUTE_GTS;
+ if (bfs)
+ flags |= ACPI_EXECUTE_BFS;
+
+ return flags;
+}
+
static u8 sleep_states[ACPI_S_STATE_COUNT];
static void acpi_sleep_tts_switch(u32 acpi_state)
@@ -243,6 +262,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
{
acpi_status status = AE_OK;
u32 acpi_state = acpi_target_sleep_state;
+ u8 flags = wake_sleep_flags();
int error;
ACPI_FLUSH_CPU_CACHE();
@@ -250,8 +270,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
switch (acpi_state) {
case ACPI_STATE_S1:
barrier();
- status = acpi_enter_sleep_state(acpi_state,
- ACPI_NO_OPTIONAL_METHODS);
+ status = acpi_enter_sleep_state(acpi_state, flags);
break;
case ACPI_STATE_S3:
@@ -266,7 +285,7 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
acpi_write_bit_register(ACPI_BITREG_SCI_ENABLE, 1);
/* Reprogram control registers and execute _BFS */
- acpi_leave_sleep_state_prep(acpi_state, ACPI_NO_OPTIONAL_METHODS);
+ acpi_leave_sleep_state_prep(acpi_state, flags);
/* ACPI 3.0 specs (P62) says that it's the responsibility
* of the OSPM to clear the status bit [ implying that the
@@ -530,27 +549,30 @@ static int acpi_hibernation_begin(void)
static int acpi_hibernation_enter(void)
{
+ u8 flags = wake_sleep_flags();
acpi_status status = AE_OK;
ACPI_FLUSH_CPU_CACHE();
/* This shouldn't return. If it returns, we have a problem */
- status = acpi_enter_sleep_state(ACPI_STATE_S4, ACPI_NO_OPTIONAL_METHODS);
+ status = acpi_enter_sleep_state(ACPI_STATE_S4, flags);
/* Reprogram control registers and execute _BFS */
- acpi_leave_sleep_state_prep(ACPI_STATE_S4, ACPI_NO_OPTIONAL_METHODS);
+ acpi_leave_sleep_state_prep(ACPI_STATE_S4, flags);
return ACPI_SUCCESS(status) ? 0 : -EFAULT;
}
static void acpi_hibernation_leave(void)
{
+ u8 flags = wake_sleep_flags();
+
/*
* If ACPI is not enabled by the BIOS and the boot kernel, we need to
* enable it here.
*/
acpi_enable();
/* Reprogram control registers and execute _BFS */
- acpi_leave_sleep_state_prep(ACPI_STATE_S4, ACPI_NO_OPTIONAL_METHODS);
+ acpi_leave_sleep_state_prep(ACPI_STATE_S4, flags);
/* Check the hardware signature */
if (facs && s4_hardware_signature != facs->hardware_signature) {
printk(KERN_EMERG "ACPI: Hardware changed while hibernated, "
@@ -771,10 +793,12 @@ static void acpi_power_off_prepare(void)
static void acpi_power_off(void)
{
+ u8 flags = wake_sleep_flags();
+
/* acpi_sleep_prepare(ACPI_STATE_S5) should have already been called */
printk(KERN_DEBUG "%s called\n", __func__);
local_irq_disable();
- acpi_enter_sleep_state(ACPI_STATE_S5, ACPI_NO_OPTIONAL_METHODS);
+ acpi_enter_sleep_state(ACPI_STATE_S5, flags);
}
/*
--
1.7.10.rc2.19.gfae9d
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox