From 1dda606c5f94b14a8f36c220d1d8844bab68a720 Mon Sep 17 00:00:00 2001 From: Alexander Graf Date: Wed, 8 Jun 2011 02:45:37 +0200 Subject: KVM: Add compat ioctl for KVM_SET_SIGNAL_MASK KVM has an ioctl to define which signal mask should be used while running inside VCPU_RUN. At least for big endian systems, this mask is different on 32-bit and 64-bit systems (though the size is identical). Add a compat wrapper that converts the mask to whatever the kernel accepts, allowing 32-bit kvm user space to set signal masks. This patch fixes qemu with --enable-io-thread on ppc64 hosts when running 32-bit user land. Signed-off-by: Alexander Graf Signed-off-by: Avi Kivity --- kernel/compat.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/compat.c b/kernel/compat.c index fc9eb093acd5..18197ae2d465 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -890,6 +890,7 @@ sigset_from_compat (sigset_t *set, compat_sigset_t *compat) case 1: set->sig[0] = compat->sig[0] | (((long)compat->sig[1]) << 32 ); } } +EXPORT_SYMBOL_GPL(sigset_from_compat); asmlinkage long compat_sys_rt_sigtimedwait (compat_sigset_t __user *uthese, -- cgit v1.2.3 From c9aaa8957f203bd6df83b002fb40b98390bed078 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Mon, 11 Jul 2011 15:28:14 -0400 Subject: KVM: Steal time implementation To implement steal time, we need the hypervisor to pass the guest information about how much time was spent running other processes outside the VM, while the vcpu had meaningful work to do - halt time does not count. This information is acquired through the run_delay field of delayacct/schedstats infrastructure, that counts time spent in a runqueue but not running. Steal time is a per-cpu information, so the traditional MSR-based infrastructure is used. A new msr, KVM_MSR_STEAL_TIME, holds the memory area address containing information about steal time This patch contains the hypervisor part of the steal time infrasructure, and can be backported independently of the guest portion. [avi, yongjie: export delayacct_on, to avoid build failures in some configs] Signed-off-by: Glauber Costa Tested-by: Eric B Munson CC: Rik van Riel CC: Jeremy Fitzhardinge CC: Peter Zijlstra CC: Anthony Liguori Signed-off-by: Yongjie Ren Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 9 +++++ arch/x86/include/asm/kvm_para.h | 4 +++ arch/x86/kvm/Kconfig | 1 + arch/x86/kvm/x86.c | 74 +++++++++++++++++++++++++++++++++++++++-- include/linux/kvm_host.h | 1 + kernel/delayacct.c | 2 ++ 6 files changed, 89 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index da6bbee878ca..59086a77ff13 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -389,6 +389,15 @@ struct kvm_vcpu_arch { unsigned int hw_tsc_khz; unsigned int time_offset; struct page *time_page; + + struct { + u64 msr_val; + u64 last_steal; + u64 accum_steal; + struct gfn_to_hva_cache stime; + struct kvm_steal_time steal; + } st; + u64 last_guest_tsc; u64 last_kernel_ns; u64 last_tsc_nsec; diff --git a/arch/x86/include/asm/kvm_para.h b/arch/x86/include/asm/kvm_para.h index 65f8bb9279e0..c484ba8e05ea 100644 --- a/arch/x86/include/asm/kvm_para.h +++ b/arch/x86/include/asm/kvm_para.h @@ -45,6 +45,10 @@ struct kvm_steal_time { __u32 pad[12]; }; +#define KVM_STEAL_ALIGNMENT_BITS 5 +#define KVM_STEAL_VALID_BITS ((-1ULL << (KVM_STEAL_ALIGNMENT_BITS + 1))) +#define KVM_STEAL_RESERVED_MASK (((1 << KVM_STEAL_ALIGNMENT_BITS) - 1 ) << 1) + #define KVM_MAX_MMU_OP_BATCH 32 #define KVM_ASYNC_PF_ENABLED (1 << 0) diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index 50f63648ce1b..99c3f0589faa 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -31,6 +31,7 @@ config KVM select KVM_ASYNC_PF select USER_RETURN_NOTIFIER select KVM_MMIO + select TASK_DELAY_ACCT ---help--- Support hosting fully virtualized guest machines using hardware virtualization extensions. You will need a fairly recent diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0b803f04bde7..c96cdc092484 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -808,12 +808,12 @@ EXPORT_SYMBOL_GPL(kvm_get_dr); * kvm-specific. Those are put in the beginning of the list. */ -#define KVM_SAVE_MSRS_BEGIN 8 +#define KVM_SAVE_MSRS_BEGIN 9 static u32 msrs_to_save[] = { MSR_KVM_SYSTEM_TIME, MSR_KVM_WALL_CLOCK, MSR_KVM_SYSTEM_TIME_NEW, MSR_KVM_WALL_CLOCK_NEW, HV_X64_MSR_GUEST_OS_ID, HV_X64_MSR_HYPERCALL, - HV_X64_MSR_APIC_ASSIST_PAGE, MSR_KVM_ASYNC_PF_EN, + HV_X64_MSR_APIC_ASSIST_PAGE, MSR_KVM_ASYNC_PF_EN, MSR_KVM_STEAL_TIME, MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP, MSR_STAR, #ifdef CONFIG_X86_64 @@ -1488,6 +1488,35 @@ static void kvmclock_reset(struct kvm_vcpu *vcpu) } } +static void accumulate_steal_time(struct kvm_vcpu *vcpu) +{ + u64 delta; + + if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED)) + return; + + delta = current->sched_info.run_delay - vcpu->arch.st.last_steal; + vcpu->arch.st.last_steal = current->sched_info.run_delay; + vcpu->arch.st.accum_steal = delta; +} + +static void record_steal_time(struct kvm_vcpu *vcpu) +{ + if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED)) + return; + + if (unlikely(kvm_read_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, + &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)))) + return; + + vcpu->arch.st.steal.steal += vcpu->arch.st.accum_steal; + vcpu->arch.st.steal.version += 2; + vcpu->arch.st.accum_steal = 0; + + kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, + &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)); +} + int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) { switch (msr) { @@ -1570,6 +1599,33 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) if (kvm_pv_enable_async_pf(vcpu, data)) return 1; break; + case MSR_KVM_STEAL_TIME: + + if (unlikely(!sched_info_on())) + return 1; + + if (data & KVM_STEAL_RESERVED_MASK) + return 1; + + if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.st.stime, + data & KVM_STEAL_VALID_BITS)) + return 1; + + vcpu->arch.st.msr_val = data; + + if (!(data & KVM_MSR_ENABLED)) + break; + + vcpu->arch.st.last_steal = current->sched_info.run_delay; + + preempt_disable(); + accumulate_steal_time(vcpu); + preempt_enable(); + + kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); + + break; + case MSR_IA32_MCG_CTL: case MSR_IA32_MCG_STATUS: case MSR_IA32_MC0_CTL ... MSR_IA32_MC0_CTL + 4 * KVM_MAX_MCE_BANKS - 1: @@ -1855,6 +1911,9 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_KVM_ASYNC_PF_EN: data = vcpu->arch.apf.msr_val; break; + case MSR_KVM_STEAL_TIME: + data = vcpu->arch.st.msr_val; + break; case MSR_IA32_P5_MC_ADDR: case MSR_IA32_P5_MC_TYPE: case MSR_IA32_MCG_CAP: @@ -2166,6 +2225,9 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) kvm_migrate_timers(vcpu); vcpu->cpu = cpu; } + + accumulate_steal_time(vcpu); + kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); } void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) @@ -2487,6 +2549,10 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, (1 << KVM_FEATURE_CLOCKSOURCE2) | (1 << KVM_FEATURE_ASYNC_PF) | (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT); + + if (sched_info_on()) + entry->eax |= (1 << KVM_FEATURE_STEAL_TIME); + entry->ebx = 0; entry->ecx = 0; entry->edx = 0; @@ -5470,6 +5536,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) r = 1; goto out; } + if (kvm_check_request(KVM_REQ_STEAL_UPDATE, vcpu)) + record_steal_time(vcpu); + } r = kvm_mmu_reload(vcpu); @@ -6206,6 +6275,7 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) kvm_make_request(KVM_REQ_EVENT, vcpu); vcpu->arch.apf.msr_val = 0; + vcpu->arch.st.msr_val = 0; kvmclock_reset(vcpu); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index f7df0a3b031d..c8e023902f79 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -47,6 +47,7 @@ #define KVM_REQ_DEACTIVATE_FPU 10 #define KVM_REQ_EVENT 11 #define KVM_REQ_APF_HALT 12 +#define KVM_REQ_STEAL_UPDATE 13 #define KVM_USERSPACE_IRQ_SOURCE_ID 0 diff --git a/kernel/delayacct.c b/kernel/delayacct.c index ead9b610aa71..418b3f7053aa 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -19,8 +19,10 @@ #include #include #include +#include int delayacct_on __read_mostly = 1; /* Delay accounting turned on/off */ +EXPORT_SYMBOL_GPL(delayacct_on); struct kmem_cache *delayacct_cache; static int __init delayacct_setup_disable(char *str) -- cgit v1.2.3 From e6e6685accfa81f509fadfc9624bc7c3862d75c4 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Mon, 11 Jul 2011 15:28:17 -0400 Subject: KVM guest: Steal time accounting This patch accounts steal time time in account_process_tick. If one or more tick is considered stolen in the current accounting cycle, user/system accounting is skipped. Idle is fine, since the hypervisor does not report steal time if the guest is halted. Accounting steal time from the core scheduler give us the advantage of direct acess to the runqueue data. In a later opportunity, it can be used to tweak cpu power and make the scheduler aware of the time it lost. [avi: doesn't exist on many archs] Signed-off-by: Glauber Costa Acked-by: Rik van Riel Acked-by: Peter Zijlstra Tested-by: Eric B Munson CC: Jeremy Fitzhardinge CC: Anthony Liguori Signed-off-by: Avi Kivity --- kernel/sched.c | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 3f2e502d609b..f98a28b19b2a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -75,6 +75,9 @@ #include #include #include +#ifdef CONFIG_PARAVIRT +#include +#endif #include "sched_cpupri.h" #include "workqueue_sched.h" @@ -528,6 +531,9 @@ struct rq { #ifdef CONFIG_IRQ_TIME_ACCOUNTING u64 prev_irq_time; #endif +#ifdef CONFIG_PARAVIRT + u64 prev_steal_time; +#endif /* calc_load related fields */ unsigned long calc_load_update; @@ -1953,6 +1959,18 @@ void account_system_vtime(struct task_struct *curr) } EXPORT_SYMBOL_GPL(account_system_vtime); +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ + +#ifdef CONFIG_PARAVIRT +static inline u64 steal_ticks(u64 steal) +{ + if (unlikely(steal > NSEC_PER_SEC)) + return div_u64(steal, TICK_NSEC); + + return __iter_div_u64_rem(steal, TICK_NSEC, &steal); +} +#endif + static void update_rq_clock_task(struct rq *rq, s64 delta) { s64 irq_delta; @@ -3845,6 +3863,25 @@ void account_idle_time(cputime_t cputime) cpustat->idle = cputime64_add(cpustat->idle, cputime64); } +static __always_inline bool steal_account_process_tick(void) +{ +#ifdef CONFIG_PARAVIRT + if (static_branch(¶virt_steal_enabled)) { + u64 steal, st = 0; + + steal = paravirt_steal_clock(smp_processor_id()); + steal -= this_rq()->prev_steal_time; + + st = steal_ticks(steal); + this_rq()->prev_steal_time += st * TICK_NSEC; + + account_steal_time(st); + return st; + } +#endif + return false; +} + #ifndef CONFIG_VIRT_CPU_ACCOUNTING #ifdef CONFIG_IRQ_TIME_ACCOUNTING @@ -3876,6 +3913,9 @@ static void irqtime_account_process_tick(struct task_struct *p, int user_tick, cputime64_t tmp = cputime_to_cputime64(cputime_one_jiffy); struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; + if (steal_account_process_tick()) + return; + if (irqtime_account_hi_update()) { cpustat->irq = cputime64_add(cpustat->irq, tmp); } else if (irqtime_account_si_update()) { @@ -3929,6 +3969,9 @@ void account_process_tick(struct task_struct *p, int user_tick) return; } + if (steal_account_process_tick()) + return; + if (user_tick) account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); else if ((p != rq->idle) || (irq_count() != HARDIRQ_OFFSET)) -- cgit v1.2.3 From 095c0aa83e52d6c3dd7168610746703921f570af Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Mon, 11 Jul 2011 15:28:18 -0400 Subject: sched: adjust scheduler cpu power for stolen time This patch makes update_rq_clock() aware of steal time. The mechanism of operation is not different from irq_time, and follows the same principles. This lives in a CONFIG option itself, and can be compiled out independently of the rest of steal time reporting. The effect of disabling it is that the scheduler will still report steal time (that cannot be disabled), but won't use this information for cpu power adjustments. Everytime update_rq_clock_task() is invoked, we query information about how much time was stolen since last call, and feed it into sched_rt_avg_update(). Although steal time reporting in account_process_tick() keeps track of the last time we read the steal clock, in prev_steal_time, this patch do it independently using another field, prev_steal_time_rq. This is because otherwise, information about time accounted in update_process_tick() would never reach us in update_rq_clock(). Signed-off-by: Glauber Costa Acked-by: Rik van Riel Acked-by: Peter Zijlstra Tested-by: Eric B Munson CC: Jeremy Fitzhardinge CC: Anthony Liguori Signed-off-by: Avi Kivity --- arch/x86/Kconfig | 12 ++++++++++++ kernel/sched.c | 47 +++++++++++++++++++++++++++++++++++++---------- kernel/sched_features.h | 4 ++-- 3 files changed, 51 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index da349723d411..1f03e221a01e 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -512,6 +512,18 @@ menuconfig PARAVIRT_GUEST if PARAVIRT_GUEST +config PARAVIRT_TIME_ACCOUNTING + bool "Paravirtual steal time accounting" + select PARAVIRT + default n + ---help--- + Select this option to enable fine granularity task steal time + accounting. Time spent executing other tasks in parallel with + the current vCPU is discounted from the vCPU power. To account for + that, there can be a small performance impact. + + If in doubt, say N here. + source "arch/x86/xen/Kconfig" config KVM_CLOCK diff --git a/kernel/sched.c b/kernel/sched.c index f98a28b19b2a..b35ac50b26c8 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -534,6 +534,9 @@ struct rq { #ifdef CONFIG_PARAVIRT u64 prev_steal_time; #endif +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + u64 prev_steal_time_rq; +#endif /* calc_load related fields */ unsigned long calc_load_update; @@ -1973,8 +1976,14 @@ static inline u64 steal_ticks(u64 steal) static void update_rq_clock_task(struct rq *rq, s64 delta) { - s64 irq_delta; - +/* + * In theory, the compile should just see 0 here, and optimize out the call + * to sched_rt_avg_update. But I don't trust it... + */ +#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) + s64 steal = 0, irq_delta = 0; +#endif +#ifdef CONFIG_IRQ_TIME_ACCOUNTING irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; /* @@ -1997,12 +2006,35 @@ static void update_rq_clock_task(struct rq *rq, s64 delta) rq->prev_irq_time += irq_delta; delta -= irq_delta; +#endif +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + if (static_branch((¶virt_steal_rq_enabled))) { + u64 st; + + steal = paravirt_steal_clock(cpu_of(rq)); + steal -= rq->prev_steal_time_rq; + + if (unlikely(steal > delta)) + steal = delta; + + st = steal_ticks(steal); + steal = st * TICK_NSEC; + + rq->prev_steal_time_rq += steal; + + delta -= steal; + } +#endif + rq->clock_task += delta; - if (irq_delta && sched_feat(NONIRQ_POWER)) - sched_rt_avg_update(rq, irq_delta); +#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) + if ((irq_delta + steal) && sched_feat(NONTASK_POWER)) + sched_rt_avg_update(rq, irq_delta + steal); +#endif } +#ifdef CONFIG_IRQ_TIME_ACCOUNTING static int irqtime_account_hi_update(void) { struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; @@ -2037,12 +2069,7 @@ static int irqtime_account_si_update(void) #define sched_clock_irqtime (0) -static void update_rq_clock_task(struct rq *rq, s64 delta) -{ - rq->clock_task += delta; -} - -#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ +#endif #include "sched_idletask.c" #include "sched_fair.c" diff --git a/kernel/sched_features.h b/kernel/sched_features.h index be40f7371ee1..ca3b025f8669 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -61,9 +61,9 @@ SCHED_FEAT(LB_BIAS, 1) SCHED_FEAT(OWNER_SPIN, 1) /* - * Decrement CPU power based on irq activity + * Decrement CPU power based on time not spent running tasks */ -SCHED_FEAT(NONIRQ_POWER, 1) +SCHED_FEAT(NONTASK_POWER, 1) /* * Queue remote wakeups on the target CPU and process them -- cgit v1.2.3