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;
Date:   Tue, 27 Mar 2018 01:18:40 -0400
From:   Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To:     Wanpeng Li <kernellwp@gmail.com>
Cc:     LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        Andrew Cooper <andrew.cooper3@citrix.com>
Subject: Re: [PATCH 2/2] KVM: VMX: Add Force Emulation Prefix for "emulate
 the next instruction"
Message-ID: <20180327051840.GF16974@char.us.oracle.com>
References: <1522116735-4861-1-git-send-email-wanpengli@tencent.com>
 <1522116735-4861-3-git-send-email-wanpengli@tencent.com>
 <20180327044020.GB16974@char.us.oracle.com>
 <20180327045558.GC16974@char.us.oracle.com>
 <CANRm+Cw7VVNmA=DvrzbUDoTQZ_O2sTQVJiu-v-fVt+e3jfB_gw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CANRm+Cw7VVNmA=DvrzbUDoTQZ_O2sTQVJiu-v-fVt+e3jfB_gw@mail.gmail.com>
User-Agent: Mutt/1.8.3 (2017-05-23)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Mar 27, 2018 at 01:03:15PM +0800, Wanpeng Li wrote:
> 2018-03-27 12:55 GMT+08:00 Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>:
> > On Tue, Mar 27, 2018 at 12:40:20AM -0400, Konrad Rzeszutek Wilk wrote:
> >> On Mon, Mar 26, 2018 at 07:12:15PM -0700, Wanpeng Li wrote:
> >> > From: Wanpeng Li <wanpengli@tencent.com>
> >> >
> >> > This patch introduces a Force Emulation Prefix (ud2a; .ascii "kvm") for
> >> > "emulate the next instruction", the codes will be executed by emulator
> >> > instead of processor, for testing purposes.
> >>
> >> Can you expand a bit ? Why do you want this in KVM in the first place?
> 
> Please refer to the original discussion(Force Emulation Prefix part).
> https://lkml.org/lkml/2018/3/22/220

That really should be part massaged in this patch as part of the description.
> 
> >> Should this be controlled by a boolean parameter?
> >
> > .. per guest. That is instead of a global one, have a per guest one?
> 
> As Paolo pointed out offline:
> > Testing without the hacks being done by emulator.flat (TLB mismatch between instructions and data).

And also this above.

> I think a global module is enough for testing.

If so, perhaps have it wrapped with #ifdef DEBUG?

No need to put code gadgets that won't be utilized 99% of time.

> 
> Regards,
> Wanpeng Li