linux-kernel - Re: [PATCH v1 03/11] KVM: x86: dynamic kvm_apic

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]

Message-ID: <20160701154354.GA5965@potion>
Date:	Fri, 1 Jul 2016 17:43:55 +0200
From:	Radim Krčmář <rkrcmar@...hat.com>
To:	Paolo Bonzini <pbonzini@...hat.com>
Cc:	Andrew Honig <ahonig@...gle.com>, linux-kernel@...r.kernel.org,
	kvm <kvm@...r.kernel.org>, "Lan, Tianyu" <tianyu.lan@...el.com>,
	Igor Mammedov <imammedo@...hat.com>,
	Jan Kiszka <jan.kiszka@....de>, Peter Xu <peterx@...hat.com>
Subject: Re: [PATCH v1 03/11] KVM: x86: dynamic kvm_apic_map

2016-07-01 17:12+0200, Paolo Bonzini:
> On 01/07/2016 17:06, Paolo Bonzini wrote:
>>>> >> > Should it?
>>> Yes, x2APIC ID cannot be changed in hardware and is initialized to the
>>> intitial APIC ID.
>>> Letting LAPIC_SET change x2APIC ID would allow scenarios where userspace
>>> reuses old VMs instead of building new ones after reconfiguration.
>>> I don't think it's a sensible use case and it it is currently broken,
>>> because we don't exit to userspace when changing APIC mode, so KVM would
>>> just set APIC ID to VCPU ID on any transition and userspace couldn't
>>> amend it.
> 
> Forgot to reply about this: letting SET_LAPIC change x2APIC IDs is nonsense.
> 
> In x2APIC mode + new capability disabled SET_LAPIC should ignore the id
> register altogether for backwards compatibility.

I'd still shift SET_LAPIC APIC ID to have internal APIC ID register in
hardware-compatible format.

> In x2APIC mode + new capability enabled it should either ignore it, or
> fail if the x2APIC ID doesn't match the VCPU id.  I suspect the latter
> is better because it would help catching the case where userspace is
> erroneously shifting the id left to bits 31-24.

Yes, I'll make it EINVAL.