Received-SPF: pass (google.com: domain of linux-kernel+bounces-187022-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Thu, 23 May 2024 13:59:49 +0800
From: Chao Gao <chao.gao@intel.com>
To: Sean Christopherson <seanjc@google.com>
CC: Paolo Bonzini <pbonzini@redhat.com>, <kvm@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, Kai Huang <kai.huang@intel.com>
Subject: Re: [PATCH v2 6/6] KVM: x86: Register "emergency disable" callbacks
 when virt is enabled
Message-ID: <Zk7bVYh0iJkX/GYs@chao-email>
References: <20240522022827.1690416-1-seanjc@google.com>
 <20240522022827.1690416-7-seanjc@google.com>
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <20240522022827.1690416-7-seanjc@google.com>
Precedence: bulk
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?rpYfzyI7Z6b8ziRxFUkBgDV4jCl5Fio0PE9hvImlsowFRMcN/qCFwyrLHVmD?=
 =?us-ascii?Q?1Sr0AjgkMuXWTWINjk5kluyFqglhjC1saSJbBmS90ZsmWmCFVUTkz3/yc+Z4?=
 =?us-ascii?Q?1ZP7ertDP4uuXKA6ZJ6Ev1LnR76n7NLD7EQMO4r3Pmz0qvLW54ZyfbZeFiku?=
 =?us-ascii?Q?sGaFIEN7bU9xUfuuD/3wgzYpcVG+Dw+L46Xk/7wD/Q5jvnIuHiRlaH5nA61i?=
 =?us-ascii?Q?ip/1Zj/nu8Pa4InP7SYETFMg1iTc/ZHlEsYuLWDNUO4R0KBf0LwTlqHmKU/6?=
 =?us-ascii?Q?+KDOlCfNTBw/pIB+7F4UgC7uK34EYQt0pL5qxhtZmD7gNEwycL4j3K8fHDb4?=
 =?us-ascii?Q?/CMM1rZ4Go3QZqo4646r8CfSUceg+mk/Trcr0a0sORW5XV+EQnl2TxDwCeUA?=
 =?us-ascii?Q?X/zIObwj9iA6HMNNDlbTUwnIH8W4wU8LsMwTC5ZRRkhvYrLhGeS25kYTCu93?=
 =?us-ascii?Q?yEIgvnEWyTgNyFabttGOPUvlSh2jGIl3o49O81q/AXd6KQK/eKoNyio9mNBr?=
 =?us-ascii?Q?wcitYySdBaB+/A068FgGEB67CA/mx4Ec77UdikjkNcuaORSe3wplc51LqJgW?=
 =?us-ascii?Q?yyZhZAe722gT0xZ/duRbD03RT6xGVJf+8DIhYrqGwUJtfjQnPwvKjGkEpcB0?=
 =?us-ascii?Q?BNEEXOr2Q9BuWJ5NgW6UCPy+yHFSwuvVzr/x+HhSza8y3p42nvRJ59+kFdE3?=
 =?us-ascii?Q?97rX2rBImP+sTc8iF+LKxts8m4DIQ/DvMUdwy9ggPiXCywyAZPHQFfOSL6lk?=
 =?us-ascii?Q?YafY4Ofpi55Kn0KaBTHSKRimUYT5GITVlweLU+TAzaKeOojxSuRQMP/FIsaq?=
 =?us-ascii?Q?3+QpfZ/r/0j1Hs2YBW5bmhW+Pem+51DuyrVAgWywqYvfGQ5DAvkVIVZJBn9+?=
 =?us-ascii?Q?b7Wu05Nk+FQGmp1oHYqkKGwiE5GtyN1jMZ9nHmiMDw2YuIklzTce46U5IWr6?=
 =?us-ascii?Q?gTopW1KN5efiRPwvkCfZ4nJ7+B0mLD4s4RDNhka6c42cDZ3SndjfwDwJfga7?=
 =?us-ascii?Q?A8QRiilymggKeTRVF7WD8O9LObTbdIdvWhBgoFwsEjLgLOAbktwZKMnQXK8/?=
 =?us-ascii?Q?SbAh2OIG3mY+VERNxG+jdmT6bfYVFHMduO6oMslo8/L/wIegNct3ES3PHBHk?=
 =?us-ascii?Q?mEEpTa3G+3KfhElsyG7GejdzlWn/iqFK/N/liMM8C1rXuWBIlU8QWHja2fAI?=
 =?us-ascii?Q?YES5CEnMG25JZv4oQBM5nolmJtYysg4DtAXNh2ncZ+A0vWJe6hiM+pbFUNRL?=
 =?us-ascii?Q?iAsFuDDSvGywAmbCaKjebq/4GM2HFVJhSlQJE0UyIvr4wq+JPr/YrXoMsKfg?=
 =?us-ascii?Q?oe81X9yr8oQ3rAR/DU9gN1M/UUauAUfeZ5Zn+m64vbACsLycopmJW0gvURCM?=
 =?us-ascii?Q?FU3IhLK5f6dTFMFNNqWPmYra58QnaU4ARPCo8usvva4AVb8GFtLe+9+uYhuQ?=
 =?us-ascii?Q?+kwTyLrVXLwHNW3rHn1xW1dhB3NsxKl+FXD27oOzx/3n3Tm2Sr5YEnEsUBsB?=
 =?us-ascii?Q?j2sNVXkPOAaK6z+JAYKO+Ih/oCXZxRn99hnDbfQ2IpI1J1ZOEr3TXCpz4e3x?=
 =?us-ascii?Q?0zYkb6Zc+wHQ6tyQMvD0dVXAuJ5iraoDgfaPRA9J?=
X-MS-Exchange-CrossTenant-Network-Message-Id: f4eec9ee-56cc-4d92-4620-08dc7aed9311
X-MS-Exchange-CrossTenant-AuthSource: CH3PR11MB8660.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 23 May 2024 05:59:57.3934
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 46c98d88-e344-4ed4-8496-4ed7712e255d
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: Da6r5BHH8sWMzs4rWYQ2vwB1xDH8XjxwlBNIZLol1OzNq7R9wLFkx0ZlDMLneRJLAY8A9Uos62VyEkExgGIDfQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: IA1PR11MB7342
X-OriginatorOrg: intel.com

On Tue, May 21, 2024 at 07:28:27PM -0700, Sean Christopherson wrote:
>Register the "disable virtualization in an emergency" callback just
>before KVM enables virtualization in hardware, as there is no functional
>need to keep the callbacks registered while KVM happens to be loaded, but
>is inactive, i.e. if KVM hasn't enabled virtualization.
>
>Note, unregistering the callback every time the last VM is destroyed could
>have measurable latency due to the synchronize_rcu() needed to ensure all
>references to the callback are dropped before KVM is unloaded.  But the
>latency should be a small fraction of the total latency of disabling
>virtualization across all CPUs, and userspace can set enable_virt_at_load
>to completely eliminate the runtime overhead.
>
>Add a pointer in kvm_x86_ops to allow vendor code to provide its callback.
>There is no reason to force vendor code to do the registration, and either
>way KVM would need a new kvm_x86_ops hook.
>
>Suggested-by: Kai Huang <kai.huang@intel.com>
>Signed-off-by: Sean Christopherson <seanjc@google.com>

Reviewed-by: Chao Gao <chao.gao@intel.com>

..

>--- a/arch/x86/kvm/vmx/vmx.c
>+++ b/arch/x86/kvm/vmx/vmx.c
>@@ -753,7 +753,7 @@ static int kvm_cpu_vmxoff(void)
> 	return -EIO;
> }
> 
>-static void vmx_emergency_disable(void)
>+void vmx_emergency_disable(void)
> {
> 	int cpu = raw_smp_processor_id();
> 	struct loaded_vmcs *v;
>@@ -8613,8 +8613,6 @@ static void __vmx_exit(void)
> {
> 	allow_smaller_maxphyaddr = false;
> 
>-	cpu_emergency_unregister_virt_callback(vmx_emergency_disable);
>-
> 	vmx_cleanup_l1d_flush();
> }
> 
>@@ -8661,8 +8659,6 @@ static int __init vmx_init(void)
> 		pi_init_cpu(cpu);
> 	}
> 
>-	cpu_emergency_register_virt_callback(vmx_emergency_disable);
>-

Nit: with the removal of calls to cpu_emergency_(un)register_virt_callback,
there is no need to include asm/reboot.h in vmx.c any more. right?