From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758107Ab2HWMaa (ORCPT ); Thu, 23 Aug 2012 08:30:30 -0400 Received: from e28smtp04.in.ibm.com ([122.248.162.4]:50128 "EHLO e28smtp04.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753356Ab2HWMa1 (ORCPT ); Thu, 23 Aug 2012 08:30:27 -0400 Message-ID: <50362257.6060006@linux.vnet.ibm.com> Date: Thu, 23 Aug 2012 20:30:15 +0800 From: Xiao Guangrong User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:14.0) Gecko/20120717 Thunderbird/14.0 MIME-Version: 1.0 To: Marcelo Tosatti CC: Avi Kivity , LKML , KVM Subject: Re: [PATCH] KVM: trace the events of mmu_notifier References: <50335A27.2070306@linux.vnet.ibm.com> <20120823092405.GB32710@amt.cnet> In-Reply-To: <20120823092405.GB32710@amt.cnet> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit x-cbid: 12082312-5564-0000-0000-000004254EAF Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/23/2012 05:24 PM, Marcelo Tosatti wrote: > On Tue, Aug 21, 2012 at 05:51:35PM +0800, Xiao Guangrong wrote: >> mmu_notifier is the interface to broadcast the mm events to KVM, the >> tracepoints introduced in this patch can trace all these events, it is >> very helpful for us to notice and fix the bug caused by mm >> >> Signed-off-by: Xiao Guangrong >> --- >> include/trace/events/kvm.h | 121 ++++++++++++++++++++++++++++++++++++++++++++ >> virt/kvm/kvm_main.c | 19 +++++++ >> 2 files changed, 140 insertions(+), 0 deletions(-) >> >> diff --git a/include/trace/events/kvm.h b/include/trace/events/kvm.h >> index 7ef9e75..a855ff9 100644 >> --- a/include/trace/events/kvm.h >> +++ b/include/trace/events/kvm.h >> @@ -309,6 +309,127 @@ TRACE_EVENT( >> >> #endif >> >> +#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER) >> +DECLARE_EVENT_CLASS(mmu_notifier_address_class, >> + >> + TP_PROTO(struct kvm *kvm, unsigned long address), >> + >> + TP_ARGS(kvm, address), >> + >> + TP_STRUCT__entry( >> + __field(struct kvm *, kvm) >> + __field(unsigned long, address) >> + ), >> + >> + TP_fast_assign( >> + __entry->kvm = kvm; >> + __entry->address = address; >> + ), >> + >> + TP_printk("kvm %p address %lx", __entry->kvm, __entry->address) > > > Perhaps the pointer was useful for debugging, but otherwise, i don't > think it should be printed. But this is the only information to identify the guest if may guest are running. > >> + TP_ARGS(kvm, address, pte), >> + >> + TP_STRUCT__entry( >> + __field(struct kvm *, kvm) >> + __field(unsigned long, address) >> + __field(unsigned long, pte) >> + ), >> + >> + TP_fast_assign( >> + __entry->kvm = kvm; >> + __entry->address = address; >> + __entry->pte = pte.pte; >> + ), >> + >> + TP_printk("kvm %p address %lx pte %lx", __entry->kvm, __entry->address, >> + __entry->pte) > > The pte bits can be spelled out? (see __print_symbolic). This tracepoint is put in kvm.h which can be used on different architectures which have different pte decode. I will try to find a way to show more readable format. > >> spin_lock(&kvm->mmu_lock); >> >> + trace_kvm_mmu_notifier_clear_flush_young(kvm, address); >> + >> young = kvm_age_hva(kvm, address); >> if (young) >> kvm_flush_remote_tlbs(kvm); >> @@ -394,6 +407,9 @@ static int kvm_mmu_notifier_test_young(struct mmu_notifier *mn, >> >> idx = srcu_read_lock(&kvm->srcu); >> spin_lock(&kvm->mmu_lock); >> + >> + trace_kvm_mmu_notifier_test_young(kvm, address); >> + > > can print young information? Okay, will do it in next version. Thanks for your review, Marcelo!