linux-kernel - Re: [PATCH] KVM: x86: add hint to skip hidden rdpkru under kvm_load_host_xsave

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:   Mon, 17 May 2021 10:55:19 -0700
From:   Dave Hansen <dave.hansen@...el.com>
To:     Sean Christopherson <seanjc@...gle.com>,
        Paolo Bonzini <pbonzini@...hat.com>
Cc:     Andy Lutomirski <luto@...nel.org>, Jon Kohler <jon@...anix.com>,
        Babu Moger <babu.moger@....com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        X86 ML <x86@...nel.org>, "H. Peter Anvin" <hpa@...or.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Yu-cheng Yu <yu-cheng.yu@...el.com>,
        Tony Luck <tony.luck@...el.com>,
        Uros Bizjak <ubizjak@...il.com>,
        Petteri Aimonen <jpa@....mail.kapsi.fi>,
        Kan Liang <kan.liang@...ux.intel.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Mike Rapoport <rppt@...nel.org>,
        Benjamin Thiel <b.thiel@...teo.de>,
        Fan Yang <Fan_Yang@...u.edu.cn>,
        Juergen Gross <jgross@...e.com>,
        Dave Jiang <dave.jiang@...el.com>,
        "Peter Zijlstra (Intel)" <peterz@...radead.org>,
        Ricardo Neri <ricardo.neri-calderon@...ux.intel.com>,
        Arvind Sankar <nivedita@...m.mit.edu>,
        LKML <linux-kernel@...r.kernel.org>,
        kvm list <kvm@...r.kernel.org>
Subject: Re: [PATCH] KVM: x86: add hint to skip hidden rdpkru under
 kvm_load_host_xsave_state

On 5/17/21 10:39 AM, Sean Christopherson wrote:
> On Mon, May 17, 2021, Paolo Bonzini wrote:
>> On 14/05/21 07:11, Andy Lutomirski wrote:
>>> I don't even want to think about what happens if a perf NMI hits and
>>> accesses host user memory while the guest PKRU is live (on VMX -- I
>>> think this can't happen on SVM).
>> This is indeed a problem, which indeed cannot happen on SVM but is there on
>> VMX.  Note that the function above is not handling all of the xstate, it's
>> handling the *XSAVE state*, that is XCR0, XSS and PKRU.  Thus the window is
>> small, but it's there.
>>
>> Is it solvable at all, without having PKRU fields in the VMCS (and without
>> masking NMIs in the LAPIC which would be too expensive)?  Dave, Sean, what
>> do you think?
> The least awful solution would be to have the NMI handler restore the host's
> PKRU.  The NMI handler would need to save/restore the register, a la CR2, but the
> whole thing could be optimized to run if and only if the NMI lands in the window
> where the guest's PKRU is loaded.

What were you thinking about?  Something like:

	*this_cpu_ptr(&need_nmi_wpkru) = 1
	// Enter Guest
	__write_pkru(vcpu->arch.pkru);
	*this_cpu_ptr(&need_nmi_wpkru) = 0

And then in the NMI handler:

	u32 pkru;

	if (*this_cpu_ptr(&need_nmi_wpkru)) {
		pkru = rdpku();
		__write_pkru(vcpu->arch.pkru);
	}
	...
	copy_*_user_nmi(... whatever ...);
	...
	if (*this_cpu_ptr(&need_nmi_wpkru))
		__write_pkru(pkru);

?

I was thinking we could just blow away PKRU without saving/restoring it
in the NMI handler, but that might clobber PKRU in the window between
need_nmi_wpkru=1 and entering the guest.

But, the save/restore seems doable especially since we can do it in C
and don't have to mess with the NMI stack or anything.