Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Vitaly Kuznetsov <vkuznets@redhat.com>
To:     Roman Kagan <rkagan@virtuozzo.com>
Cc:     "kvm\@vger.kernel.org" <kvm@vger.kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        "linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "K. Y. Srinivasan" <kys@microsoft.com>,
        Haiyang Zhang <haiyangz@microsoft.com>,
        Stephen Hemminger <sthemmin@microsoft.com>,
        "x86\@kernel.org" <x86@kernel.org>,
        "Michael Kelley \(EOSG\)" <Michael.H.Kelley@microsoft.com>
Subject: Re: [PATCH v2 3/4] x86/kvm/hyper-v: direct mode for synthetic timers
In-Reply-To: <20181210132134.GA5888@rkaganb.sw.ru>
References: <20181126154732.23025-1-vkuznets@redhat.com> <20181126154732.23025-4-vkuznets@redhat.com> <20181210120637.GA13294@rkaganb.sw.ru> <87va41czk5.fsf@vitty.brq.redhat.com> <20181210132134.GA5888@rkaganb.sw.ru>
Date:   Mon, 10 Dec 2018 15:53:54 +0100
Message-ID: <87sgz5cu0t.fsf@vitty.brq.redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Roman Kagan <rkagan@virtuozzo.com> writes:

> On Mon, Dec 10, 2018 at 01:54:18PM +0100, Vitaly Kuznetsov wrote:
>> Roman Kagan <rkagan@virtuozzo.com> writes:
>> > Just noticed that the patch seems to assume that "direct" timers are
>> > allowed to use any vectors including 0-15.  I guess this is incorrect,
>> > and instead stimer_set_config should error out on direct mode with a
>> > vector less than HV_SYNIC_FIRST_VALID_VECTOR.
>> 
>> The spec is really vague about this and I'm not sure that this has
>> anything to do with HV_SYNIC_FIRST_VALID_VECTOR (as these are actually
>> not "synic" vectors, I *think* that SynIC doesn't even need to be
>> enabled to make them work).
>> 
>> I checked and Hyper-V 2016 uses vector '0xff', not sure if it proves
>> your point :-)
>> 
>> Do you envision any issues in KVM if we keep allowing vectors <
>> HV_SYNIC_FIRST_VALID_VECTOR?
>
> It's actually lapic that treats vectors 0..15 as illegal.  Nothing
> Hyper-V specific here.

Oh, right you are,

Intel SDM 10.5.2 "Valid Interrupt Vectors" says:

"The Intel 64 and IA-32 architectures define 256 vector numbers, ranging
from 0 through 255 (see Section 6.2, “Exception and Interrupt
Vectors”). Local and I/O APICs support 240 of these vectors (in the
range of 16 to 255) as valid interrupts.

When an interrupt vector in the range of 0 to 15 is sent or received
through the local APIC, the APIC indicates an illegal vector in its
Error Status Register (see Section 10.5.3, “Error Handling”). The Intel
64 and IA-32 architectures reserve vectors 16 through 31 for predefined
interrupts, exceptions, and Intel-reserved encodings (see Table
6-1). However, the local APIC does not treat vectors in this range as
illegal."

Out of pure curiosity I checked what Hyper-V does by hacking up linux
and I got "unchecked MSR access error: WRMSR to 0x400000b0" so we know
they follow the spec.

I'll send a patch to fix this, thanks!

-- 
Vitaly