linux-kernel - Re: [PATCH v18 007/121] KVM: VMX: Reorder vmx initialization with kvm vendor initialization

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Thu, 1 Feb 2024 17:34:44 +0800
From: Xiaoyao Li <xiaoyao.li@...el.com>
To: isaku.yamahata@...el.com, kvm@...r.kernel.org,
 linux-kernel@...r.kernel.org
Cc: isaku.yamahata@...il.com, Paolo Bonzini <pbonzini@...hat.com>,
 erdemaktas@...gle.com, Sean Christopherson <seanjc@...gle.com>,
 Sagi Shahar <sagis@...gle.com>, Kai Huang <kai.huang@...el.com>,
 chen.bo@...el.com, hang.yuan@...el.com, tina.zhang@...el.com
Subject: Re: [PATCH v18 007/121] KVM: VMX: Reorder vmx initialization with kvm
 vendor initialization

On 1/23/2024 7:52 AM, isaku.yamahata@...el.com wrote:
> From: Isaku Yamahata <isaku.yamahata@...el.com>
> 
> To match vmx_exit cleanup.  Now vmx_init() is before kvm_x86_vendor_init(),
> vmx_init() can initialize loaded_vmcss_on_cpu.  Oppertunistically move it
> back into vmx_init().
> 
> Signed-off-by: Isaku Yamahata <isaku.yamahata@...el.com>
> ---
> v18:
> - move the loaded_vmcss_on_cpu initialization to vmx_init().
> - fix error path of vt_init(). by Chao and Binbin
> ---
>   arch/x86/kvm/vmx/main.c    | 17 +++++++----------
>   arch/x86/kvm/vmx/vmx.c     |  6 ++++--
>   arch/x86/kvm/vmx/x86_ops.h |  2 --
>   3 files changed, 11 insertions(+), 14 deletions(-)
> 
> diff --git a/arch/x86/kvm/vmx/main.c b/arch/x86/kvm/vmx/main.c
> index 18cecf12c7c8..443db8ec5cd5 100644
> --- a/arch/x86/kvm/vmx/main.c
> +++ b/arch/x86/kvm/vmx/main.c
> @@ -171,7 +171,7 @@ struct kvm_x86_init_ops vt_init_ops __initdata = {
>   static int __init vt_init(void)
>   {
>   	unsigned int vcpu_size, vcpu_align;
> -	int cpu, r;
> +	int r;
>   
>   	if (!kvm_is_vmx_supported())
>   		return -EOPNOTSUPP;
> @@ -182,18 +182,14 @@ static int __init vt_init(void)
>   	 */
>   	hv_init_evmcs();
>   
> -	/* vmx_hardware_disable() accesses loaded_vmcss_on_cpu. */
> -	for_each_possible_cpu(cpu)
> -		INIT_LIST_HEAD(&per_cpu(loaded_vmcss_on_cpu, cpu));
> -
> -	r = kvm_x86_vendor_init(&vt_init_ops);
> -	if (r)
> -		return r;
> -
>   	r = vmx_init();
>   	if (r)
>   		goto err_vmx_init;
>   
> +	r = kvm_x86_vendor_init(&vt_init_ops);
> +	if (r)
> +		goto err_vendor_init;
> +

we cannot simply change the calling order of vmx_init() and 
kvm_x86_vendor_init(). There is dependency between them.

e.g.,

kvm_x86_vendor_init()
   -> ops->hardware_setup()
	-> vmx_hardware_setup()

will update 'enable_ept' based on hardware capability (e.g., if the 
hardware support EPT or not), while 'enable_ept' is used in vmx_init().