From fa86ee90eb1111267de67cb4272b5ce711f18cbb Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Wed, 3 Jul 2019 20:51:25 -0300
Subject: add cpuidle-haltpoll driver
Add a cpuidle driver that calls the architecture default_idle routine.
To be used in conjunction with the haltpoll governor.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
arch/x86/kernel/process.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
(limited to 'arch/x86')
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 75fea0d48c0e..5e94c4354d4e 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -580,7 +580,7 @@ void __cpuidle default_idle(void)
safe_halt();
trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, smp_processor_id());
}
-#ifdef CONFIG_APM_MODULE
+#if defined(CONFIG_APM_MODULE) || defined(CONFIG_HALTPOLL_CPUIDLE_MODULE)
EXPORT_SYMBOL(default_idle);
#endif
--
cgit
From a1c4423b02b2121108e3ea9580741e0f26309a48 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Wed, 3 Jul 2019 20:51:29 -0300
Subject: cpuidle-haltpoll: disable host side polling when kvm virtualized
When performing guest side polling, it is not necessary to
also perform host side polling.
So disable host side polling, via the new MSR interface,
when loading cpuidle-haltpoll driver.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
arch/x86/Kconfig | 7 ++++++
arch/x86/include/asm/cpuidle_haltpoll.h | 8 +++++++
arch/x86/kernel/kvm.c | 42 +++++++++++++++++++++++++++++++++
drivers/cpuidle/cpuidle-haltpoll.c | 9 ++++++-
include/linux/cpuidle_haltpoll.h | 16 +++++++++++++
5 files changed, 81 insertions(+), 1 deletion(-)
create mode 100644 arch/x86/include/asm/cpuidle_haltpoll.h
create mode 100644 include/linux/cpuidle_haltpoll.h
(limited to 'arch/x86')
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 222855cc0158..05e78acb187c 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -794,6 +794,7 @@ config KVM_GUEST
bool "KVM Guest support (including kvmclock)"
depends on PARAVIRT
select PARAVIRT_CLOCK
+ select ARCH_CPUIDLE_HALTPOLL
default y
---help---
This option enables various optimizations for running under the KVM
@@ -802,6 +803,12 @@ config KVM_GUEST
underlying device model, the host provides the guest with
timing infrastructure such as time of day, and system time
+config ARCH_CPUIDLE_HALTPOLL
+ def_bool n
+ prompt "Disable host haltpoll when loading haltpoll driver"
+ help
+ If virtualized under KVM, disable host haltpoll.
+
config PVH
bool "Support for running PVH guests"
---help---
diff --git a/arch/x86/include/asm/cpuidle_haltpoll.h b/arch/x86/include/asm/cpuidle_haltpoll.h
new file mode 100644
index 000000000000..ff8607d81526
--- /dev/null
+++ b/arch/x86/include/asm/cpuidle_haltpoll.h
@@ -0,0 +1,8 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ARCH_HALTPOLL_H
+#define _ARCH_HALTPOLL_H
+
+void arch_haltpoll_enable(void);
+void arch_haltpoll_disable(void);
+
+#endif
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index b7f34fe2171e..f48401be8ce0 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -875,3 +875,45 @@ void __init kvm_spinlock_init(void)
}
#endif /* CONFIG_PARAVIRT_SPINLOCKS */
+
+#ifdef CONFIG_ARCH_CPUIDLE_HALTPOLL
+
+static void kvm_disable_host_haltpoll(void *i)
+{
+ wrmsrl(MSR_KVM_POLL_CONTROL, 0);
+}
+
+static void kvm_enable_host_haltpoll(void *i)
+{
+ wrmsrl(MSR_KVM_POLL_CONTROL, 1);
+}
+
+void arch_haltpoll_enable(void)
+{
+ if (!kvm_para_has_feature(KVM_FEATURE_POLL_CONTROL)) {
+ printk(KERN_ERR "kvm: host does not support poll control\n");
+ printk(KERN_ERR "kvm: host upgrade recommended\n");
+ return;
+ }
+
+ preempt_disable();
+ /* Enable guest halt poll disables host halt poll */
+ kvm_disable_host_haltpoll(NULL);
+ smp_call_function(kvm_disable_host_haltpoll, NULL, 1);
+ preempt_enable();
+}
+EXPORT_SYMBOL_GPL(arch_haltpoll_enable);
+
+void arch_haltpoll_disable(void)
+{
+ if (!kvm_para_has_feature(KVM_FEATURE_POLL_CONTROL))
+ return;
+
+ preempt_disable();
+ /* Enable guest halt poll disables host halt poll */
+ kvm_enable_host_haltpoll(NULL);
+ smp_call_function(kvm_enable_host_haltpoll, NULL, 1);
+ preempt_enable();
+}
+EXPORT_SYMBOL_GPL(arch_haltpoll_disable);
+#endif
diff --git a/drivers/cpuidle/cpuidle-haltpoll.c b/drivers/cpuidle/cpuidle-haltpoll.c
index 35cfb53e9287..9ac093dcbb01 100644
--- a/drivers/cpuidle/cpuidle-haltpoll.c
+++ b/drivers/cpuidle/cpuidle-haltpoll.c
@@ -15,6 +15,7 @@
#include <linux/module.h>
#include <linux/sched/idle.h>
#include <linux/kvm_para.h>
+#include <linux/cpuidle_haltpoll.h>
static int default_enter_idle(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index)
@@ -47,6 +48,7 @@ static struct cpuidle_driver haltpoll_driver = {
static int __init haltpoll_init(void)
{
+ int ret;
struct cpuidle_driver *drv = &haltpoll_driver;
cpuidle_poll_state_init(drv);
@@ -54,11 +56,16 @@ static int __init haltpoll_init(void)
if (!kvm_para_available())
return 0;
- return cpuidle_register(&haltpoll_driver, NULL);
+ ret = cpuidle_register(&haltpoll_driver, NULL);
+ if (ret == 0)
+ arch_haltpoll_enable();
+
+ return ret;
}
static void __exit haltpoll_exit(void)
{
+ arch_haltpoll_disable();
cpuidle_unregister(&haltpoll_driver);
}
diff --git a/include/linux/cpuidle_haltpoll.h b/include/linux/cpuidle_haltpoll.h
new file mode 100644
index 000000000000..fe5954c2409e
--- /dev/null
+++ b/include/linux/cpuidle_haltpoll.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _CPUIDLE_HALTPOLL_H
+#define _CPUIDLE_HALTPOLL_H
+
+#ifdef CONFIG_ARCH_CPUIDLE_HALTPOLL
+#include <asm/cpuidle_haltpoll.h>
+#else
+static inline void arch_haltpoll_enable(void)
+{
+}
+
+static inline void arch_haltpoll_disable(void)
+{
+}
+#endif
+#endif
--
cgit
From 97d3eb9da84cae0548359b0aecb8619faad003b7 Mon Sep 17 00:00:00 2001
From: Joao Martins <joao.m.martins@oracle.com>
Date: Mon, 2 Sep 2019 11:40:31 +0100
Subject: cpuidle-haltpoll: vcpu hotplug support
When cpus != maxcpus cpuidle-haltpoll will fail to register all vcpus
past the online ones and thus fail to register the idle driver.
This is because cpuidle_add_sysfs() will return with -ENODEV as a
consequence from get_cpu_device() return no device for a non-existing
CPU.
Instead switch to cpuidle_register_driver() and manually register each
of the present cpus through cpuhp_setup_state() callbacks and future
ones that get onlined or offlined. This mimmics similar logic that
intel_idle does.
Fixes: fa86ee90eb11 ("add cpuidle-haltpoll driver")
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
arch/x86/include/asm/cpuidle_haltpoll.h | 4 +-
arch/x86/kernel/kvm.c | 18 +++------
drivers/cpuidle/cpuidle-haltpoll.c | 68 ++++++++++++++++++++++++++++++---
include/linux/cpuidle_haltpoll.h | 4 +-
4 files changed, 73 insertions(+), 21 deletions(-)
(limited to 'arch/x86')
diff --git a/arch/x86/include/asm/cpuidle_haltpoll.h b/arch/x86/include/asm/cpuidle_haltpoll.h
index ff8607d81526..c8b39c6716ff 100644
--- a/arch/x86/include/asm/cpuidle_haltpoll.h
+++ b/arch/x86/include/asm/cpuidle_haltpoll.h
@@ -2,7 +2,7 @@
#ifndef _ARCH_HALTPOLL_H
#define _ARCH_HALTPOLL_H
-void arch_haltpoll_enable(void);
-void arch_haltpoll_disable(void);
+void arch_haltpoll_enable(unsigned int cpu);
+void arch_haltpoll_disable(unsigned int cpu);
#endif
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index f48401be8ce0..60bab4a3b36b 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -888,32 +888,26 @@ static void kvm_enable_host_haltpoll(void *i)
wrmsrl(MSR_KVM_POLL_CONTROL, 1);
}
-void arch_haltpoll_enable(void)
+void arch_haltpoll_enable(unsigned int cpu)
{
if (!kvm_para_has_feature(KVM_FEATURE_POLL_CONTROL)) {
- printk(KERN_ERR "kvm: host does not support poll control\n");
- printk(KERN_ERR "kvm: host upgrade recommended\n");
+ pr_err_once("kvm: host does not support poll control\n");
+ pr_err_once("kvm: host upgrade recommended\n");
return;
}
- preempt_disable();
/* Enable guest halt poll disables host halt poll */
- kvm_disable_host_haltpoll(NULL);
- smp_call_function(kvm_disable_host_haltpoll, NULL, 1);
- preempt_enable();
+ smp_call_function_single(cpu, kvm_disable_host_haltpoll, NULL, 1);
}
EXPORT_SYMBOL_GPL(arch_haltpoll_enable);
-void arch_haltpoll_disable(void)
+void arch_haltpoll_disable(unsigned int cpu)
{
if (!kvm_para_has_feature(KVM_FEATURE_POLL_CONTROL))
return;
- preempt_disable();
/* Enable guest halt poll disables host halt poll */
- kvm_enable_host_haltpoll(NULL);
- smp_call_function(kvm_enable_host_haltpoll, NULL, 1);
- preempt_enable();
+ smp_call_function_single(cpu, kvm_enable_host_haltpoll, NULL, 1);
}
EXPORT_SYMBOL_GPL(arch_haltpoll_disable);
#endif
diff --git a/drivers/cpuidle/cpuidle-haltpoll.c b/drivers/cpuidle/cpuidle-haltpoll.c
index 9ac093dcbb01..56d8ab814466 100644
--- a/drivers/cpuidle/cpuidle-haltpoll.c
+++ b/drivers/cpuidle/cpuidle-haltpoll.c
@@ -11,12 +11,16 @@
*/
#include <linux/init.h>
+#include <linux/cpu.h>
#include <linux/cpuidle.h>
#include <linux/module.h>
#include <linux/sched/idle.h>
#include <linux/kvm_para.h>
#include <linux/cpuidle_haltpoll.h>
+static struct cpuidle_device __percpu *haltpoll_cpuidle_devices;
+static enum cpuhp_state haltpoll_hp_state;
+
static int default_enter_idle(struct cpuidle_device *dev,
struct cpuidle_driver *drv, int index)
{
@@ -46,6 +50,46 @@ static struct cpuidle_driver haltpoll_driver = {
.state_count = 2,
};
+static int haltpoll_cpu_online(unsigned int cpu)
+{
+ struct cpuidle_device *dev;
+
+ dev = per_cpu_ptr(haltpoll_cpuidle_devices, cpu);
+ if (!dev->registered) {
+ dev->cpu = cpu;
+ if (cpuidle_register_device(dev)) {
+ pr_notice("cpuidle_register_device %d failed!\n", cpu);
+ return -EIO;
+ }
+ arch_haltpoll_enable(cpu);
+ }
+
+ return 0;
+}
+
+static int haltpoll_cpu_offline(unsigned int cpu)
+{
+ struct cpuidle_device *dev;
+
+ dev = per_cpu_ptr(haltpoll_cpuidle_devices, cpu);
+ if (dev->registered) {
+ arch_haltpoll_disable(cpu);
+ cpuidle_unregister_device(dev);
+ }
+
+ return 0;
+}
+
+static void haltpoll_uninit(void)
+{
+ if (haltpoll_hp_state)
+ cpuhp_remove_state(haltpoll_hp_state);
+ cpuidle_unregister_driver(&haltpoll_driver);
+
+ free_percpu(haltpoll_cpuidle_devices);
+ haltpoll_cpuidle_devices = NULL;
+}
+
static int __init haltpoll_init(void)
{
int ret;
@@ -56,17 +100,31 @@ static int __init haltpoll_init(void)
if (!kvm_para_available())
return 0;
- ret = cpuidle_register(&haltpoll_driver, NULL);
- if (ret == 0)
- arch_haltpoll_enable();
+ ret = cpuidle_register_driver(drv);
+ if (ret < 0)
+ return ret;
+
+ haltpoll_cpuidle_devices = alloc_percpu(struct cpuidle_device);
+ if (haltpoll_cpuidle_devices == NULL) {
+ cpuidle_unregister_driver(drv);
+ return -ENOMEM;
+ }
+
+ ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "cpuidle/haltpoll:online",
+ haltpoll_cpu_online, haltpoll_cpu_offline);
+ if (ret < 0) {
+ haltpoll_uninit();
+ } else {
+ haltpoll_hp_state = ret;
+ ret = 0;
+ }
return ret;
}
static void __exit haltpoll_exit(void)
{
- arch_haltpoll_disable();
- cpuidle_unregister(&haltpoll_driver);
+ haltpoll_uninit();
}
module_init(haltpoll_init);
diff --git a/include/linux/cpuidle_haltpoll.h b/include/linux/cpuidle_haltpoll.h
index fe5954c2409e..d50c1e0411a2 100644
--- a/include/linux/cpuidle_haltpoll.h
+++ b/include/linux/cpuidle_haltpoll.h
@@ -5,11 +5,11 @@
#ifdef CONFIG_ARCH_CPUIDLE_HALTPOLL
#include <asm/cpuidle_haltpoll.h>
#else
-static inline void arch_haltpoll_enable(void)
+static inline void arch_haltpoll_enable(unsigned int cpu)
{
}
-static inline void arch_haltpoll_disable(void)
+static inline void arch_haltpoll_disable(unsigned int cpu)
{
}
#endif
--
cgit
From 1328edca4a142ee3c7442d1eece2c3ca383eff35 Mon Sep 17 00:00:00 2001
From: Wanpeng Li <wanpengli@tencent.com>
Date: Thu, 29 Aug 2019 16:49:57 +0800
Subject: cpuidle-haltpoll: Enable kvm guest polling when dedicated physical
CPUs are available
The downside of guest side polling is that polling is performed even
with other runnable tasks in the host. However, even if poll in kvm
can aware whether or not other runnable tasks in the same pCPU, it
can still incur extra overhead in over-subscribe scenario. Now we can
just enable guest polling when dedicated pCPUs are available.
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
arch/x86/kernel/kvm.c | 1 +
drivers/cpuidle/cpuidle-haltpoll.c | 3 ++-
2 files changed, 3 insertions(+), 1 deletion(-)
(limited to 'arch/x86')
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 60bab4a3b36b..3726ba48f763 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -711,6 +711,7 @@ unsigned int kvm_arch_para_hints(void)
{
return cpuid_edx(kvm_cpuid_base() | KVM_CPUID_FEATURES);
}
+EXPORT_SYMBOL_GPL(kvm_arch_para_hints);
static uint32_t __init kvm_detect(void)
{
diff --git a/drivers/cpuidle/cpuidle-haltpoll.c b/drivers/cpuidle/cpuidle-haltpoll.c
index 49a65c6fe91e..932390b028f1 100644
--- a/drivers/cpuidle/cpuidle-haltpoll.c
+++ b/drivers/cpuidle/cpuidle-haltpoll.c
@@ -97,7 +97,8 @@ static int __init haltpoll_init(void)
cpuidle_poll_state_init(drv);
- if (!kvm_para_available())
+ if (!kvm_para_available() ||
+ !kvm_para_has_hint(KVM_HINTS_REALTIME))
return -ENODEV;
ret = cpuidle_register_driver(drv);
--
cgit