From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758261Ab1GKThK (ORCPT ); Mon, 11 Jul 2011 15:37:10 -0400 Received: from mx1.redhat.com ([209.132.183.28]:28508 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757896Ab1GKThB (ORCPT ); Mon, 11 Jul 2011 15:37:01 -0400 From: Glauber Costa To: kvm@vger.kernel.org Cc: linux-kernel@vger.kernel.org, Jeremy Fitzhardinge , Avi Kivity , Anthony Liguori Subject: [PATCH v6 7/9] KVM-GST: KVM Steal time accounting Date: Mon, 11 Jul 2011 15:28:17 -0400 Message-Id: <1310412499-3873-8-git-send-email-glommer@redhat.com> In-Reply-To: <1310412499-3873-1-git-send-email-glommer@redhat.com> References: <1310412499-3873-1-git-send-email-glommer@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 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. Signed-off-by: Glauber Costa Acked-by: Rik van Riel Acked-by: Peter Zijlstra Tested-by: Eric B Munson CC: Jeremy Fitzhardinge CC: Avi Kivity CC: Anthony Liguori --- kernel/sched.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 files changed, 41 insertions(+), 0 deletions(-) diff --git a/kernel/sched.c b/kernel/sched.c index 3f2e502..aa6c030 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -75,6 +75,7 @@ #include #include #include +#include #include "sched_cpupri.h" #include "workqueue_sched.h" @@ -528,6 +529,9 @@ 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 +1957,18 @@ 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 +3861,25 @@ 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 +3911,9 @@ 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 +3967,9 @@ 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)) -- 1.7.3.4