linux-kernel - Re: [PATCH 1/2] KVM: KVM-on-hyperv: shorten no-entry section on reenlightenment

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:   Thu, 12 Aug 2021 17:53:17 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Marcelo Tosatti <mtosatti@...hat.com>
Cc:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
        vkuznets@...hat.com
Subject: Re: [PATCH 1/2] KVM: KVM-on-hyperv: shorten no-entry section on
 reenlightenment

On 11/08/21 20:03, Marcelo Tosatti wrote:
>          hyperv_stop_tsc_emulation();
> 
>          /* TSC frequency always matches when on Hyper-V */
>          for_each_present_cpu(cpu)
>                  per_cpu(cpu_tsc_khz, cpu) = tsc_khz;
>          kvm_max_guest_tsc_khz = tsc_khz;

Yeah, it's more complicated than this.  The right sequence is:

- update the master clock

- update the TSC page parameters

- stop TSC emulation

There is no need to invalidate the TSC page.

Related to this, after kvm_hv_invalidate_tsc_page the sequence value in 
the Hyper-V TSC page will always be 1, which is wrong.  I'll take a look 
at that too.

Paolo