linux-kernel - Re: [PATCH 07/10] KVM: VMX: Introduce test mode related to EPT violation VE

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] [day] [month] [year] [list]

Date: Tue, 16 Apr 2024 19:52:08 +0200
From: Paolo Bonzini <pbonzini@...hat.com>
To: Chao Gao <chao.gao@...el.com>
Cc: linux-kernel@...r.kernel.org, kvm@...r.kernel.org, 
	Isaku Yamahata <isaku.yamahata@...el.com>
Subject: Re: [PATCH 07/10] KVM: VMX: Introduce test mode related to EPT
 violation VE

On Mon, Apr 15, 2024 at 3:22 PM Chao Gao <chao.gao@...el.com> wrote:
>
> >
> >-      if (cpu_has_secondary_exec_ctrls())
> >+      if (cpu_has_secondary_exec_ctrls()) {
> >               secondary_exec_controls_set(vmx, vmx_secondary_exec_control(vmx));
> >+              if (secondary_exec_controls_get(vmx) &
> >+                  SECONDARY_EXEC_EPT_VIOLATION_VE) {
> >+                      if (!vmx->ve_info) {
>
> how about allocating ve_info in vmx_vcpu_create()? It is better to me because:
>
> a. symmetry. ve_info is free'd in vmx_vcpu_free().
> b. no need to check if this is the first call of init_vmcs(). and ENOMEM can
> be returned on allocation failure.

There is no need to return ENOMEM however, it is okay to disable the test.

However I agree that doing it in vmx_vcpu_create(), conditional on
vmcs_config.cpu_based_2nd_exec_ctrl, is a bit cleaner.

Paolo