From patchwork Wed Nov 21 02:26:35 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yanfei Zhang X-Patchwork-Id: 1777091 Return-Path: X-Original-To: patchwork-kvm@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork2.kernel.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by patchwork2.kernel.org (Postfix) with ESMTP id 2EAA8DFF38 for ; Wed, 21 Nov 2012 02:28:50 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754256Ab2KUC23 (ORCPT ); Tue, 20 Nov 2012 21:28:29 -0500 Received: from cn.fujitsu.com ([222.73.24.84]:40599 "EHLO song.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1753845Ab2KUC20 (ORCPT ); Tue, 20 Nov 2012 21:28:26 -0500 X-IronPort-AV: E=Sophos;i="4.83,290,1352044800"; d="scan'208";a="6242751" Received: from unknown (HELO tang.cn.fujitsu.com) ([10.167.250.3]) by song.cn.fujitsu.com with ESMTP; 21 Nov 2012 10:26:42 +0800 Received: from fnstmail02.fnst.cn.fujitsu.com (tang.cn.fujitsu.com [127.0.0.1]) by tang.cn.fujitsu.com (8.14.3/8.13.1) with ESMTP id qAL2SOcY024133; Wed, 21 Nov 2012 10:28:24 +0800 Received: from [10.167.226.121] ([10.167.226.121]) by fnstmail02.fnst.cn.fujitsu.com (Lotus Domino Release 8.5.3) with ESMTP id 2012112110280903-1271577 ; Wed, 21 Nov 2012 10:28:09 +0800 Message-ID: <50AC3BDB.5070007@cn.fujitsu.com> Date: Wed, 21 Nov 2012 10:26:35 +0800 From: Zhang Yanfei User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.8) Gecko/20121012 Thunderbird/10.0.8 MIME-Version: 1.0 To: "x86@kernel.org" , "kexec@lists.infradead.org" , Marcelo Tosatti , Gleb Natapov CC: "linux-kernel@vger.kernel.org" , "kvm@vger.kernel.org" Subject: [PATCH v5 2/2] KVM-INTEL: add a notifier and a bitmap to support VMCLEAR in kdump References: <50AC3AA9.7080004@cn.fujitsu.com> In-Reply-To: <50AC3AA9.7080004@cn.fujitsu.com> X-MIMETrack: Itemize by SMTP Server on mailserver/fnst(Release 8.5.3|September 15, 2011) at 2012/11/21 10:28:09, Serialize by Router on mailserver/fnst(Release 8.5.3|September 15, 2011) at 2012/11/21 10:28:09, Serialize complete at 2012/11/21 10:28:09 Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org The notifier will be registered in crash_notifier_list when loading kvm-intel module. And the bitmap indicates whether we should do VMCLEAR operation in kdump. The bits in the bitmap are set/unset according to different conditions. Signed-off-by: Zhang Yanfei --- arch/x86/kvm/vmx.c | 85 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 85 insertions(+), 0 deletions(-) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 4ff0ab9..3bbdd75 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -41,6 +41,7 @@ #include #include #include +#include #include "trace.h" @@ -963,6 +964,30 @@ static void vmcs_load(struct vmcs *vmcs) vmcs, phys_addr); } +#ifdef CONFIG_KEXEC +/* + * This bitmap is used to indicate whether the vmclear + * operation is enabled on all cpus. All disabled by + * default. + */ +static cpumask_t crash_vmclear_enabled_bitmap = CPU_MASK_NONE; + +static inline void crash_enable_local_vmclear(int cpu) +{ + cpumask_set_cpu(cpu, &crash_vmclear_enabled_bitmap); +} + +static inline void crash_disable_local_vmclear(int cpu) +{ + cpumask_clear_cpu(cpu, &crash_vmclear_enabled_bitmap); +} + +static inline int crash_local_vmclear_enabled(int cpu) +{ + return cpumask_test_cpu(cpu, &crash_vmclear_enabled_bitmap); +} +#endif + static void __loaded_vmcs_clear(void *arg) { struct loaded_vmcs *loaded_vmcs = arg; @@ -972,8 +997,14 @@ static void __loaded_vmcs_clear(void *arg) return; /* vcpu migration can race with cpu offline */ if (per_cpu(current_vmcs, cpu) == loaded_vmcs->vmcs) per_cpu(current_vmcs, cpu) = NULL; +#ifdef CONFIG_KEXEC + crash_disable_local_vmclear(cpu); +#endif list_del(&loaded_vmcs->loaded_vmcss_on_cpu_link); loaded_vmcs_init(loaded_vmcs); +#ifdef CONFIG_KEXEC + crash_enable_local_vmclear(cpu); +#endif } static void loaded_vmcs_clear(struct loaded_vmcs *loaded_vmcs) @@ -1491,8 +1522,14 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu) kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu); local_irq_disable(); +#ifdef CONFIG_KEXEC + crash_disable_local_vmclear(cpu); +#endif list_add(&vmx->loaded_vmcs->loaded_vmcss_on_cpu_link, &per_cpu(loaded_vmcss_on_cpu, cpu)); +#ifdef CONFIG_KEXEC + crash_enable_local_vmclear(cpu); +#endif local_irq_enable(); /* @@ -2302,6 +2339,20 @@ static int hardware_enable(void *garbage) return -EBUSY; INIT_LIST_HEAD(&per_cpu(loaded_vmcss_on_cpu, cpu)); + +#ifdef CONFIG_KEXEC + /* + * Now we can enable the vmclear operation in kdump + * since the loaded_vmcss_on_cpu list on this cpu + * has been initialized. + * + * Though the cpu is not in VMX operation now, there + * is no problem to enable the vmclear operation + * for the loaded_vmcss_on_cpu list is empty! + */ + crash_enable_local_vmclear(cpu); +#endif + rdmsrl(MSR_IA32_FEATURE_CONTROL, old); test_bits = FEATURE_CONTROL_LOCKED; @@ -2335,6 +2386,22 @@ static void vmclear_local_loaded_vmcss(void) __loaded_vmcs_clear(v); } +#ifdef CONFIG_KEXEC +static int crash_vmclear_local_loaded_vmcss(struct notifier_block *this, + unsigned long val, void *ptr) +{ + int cpu = raw_smp_processor_id(); + + if (crash_local_vmclear_enabled(cpu)) + vmclear_local_loaded_vmcss(); + + return NOTIFY_DONE; +} + +static struct notifier_block crash_vmclear_notifier = { + .notifier_call = crash_vmclear_local_loaded_vmcss, +}; +#endif /* Just like cpu_vmxoff(), but with the __kvm_handle_fault_on_reboot() * tricks. @@ -2348,6 +2415,14 @@ static void hardware_disable(void *garbage) { if (vmm_exclusive) { vmclear_local_loaded_vmcss(); +#ifdef CONFIG_KEXEC + /* + * vmclear operation in kdump should be disabled here + * because the cpu is going to exit VMX operation + * and the loaded_vmcss_on_cpu list may not be empty! + */ + crash_disable_local_vmclear(raw_smp_processor_id()); +#endif kvm_cpu_vmxoff(); } write_cr4(read_cr4() & ~X86_CR4_VMXE); @@ -7230,6 +7305,11 @@ static int __init vmx_init(void) if (r) goto out3; +#ifdef CONFIG_KEXEC + atomic_notifier_chain_register(&crash_notifier_list, + &crash_vmclear_notifier); +#endif + vmx_disable_intercept_for_msr(MSR_FS_BASE, false); vmx_disable_intercept_for_msr(MSR_GS_BASE, false); vmx_disable_intercept_for_msr(MSR_KERNEL_GS_BASE, true); @@ -7265,6 +7345,11 @@ static void __exit vmx_exit(void) free_page((unsigned long)vmx_io_bitmap_b); free_page((unsigned long)vmx_io_bitmap_a); +#ifdef CONFIG_KEXEC + atomic_notifier_chain_unregister(&crash_notifier_list, + &crash_vmclear_notifier); +#endif + kvm_exit(); }