linux-kernel - Re: [PATCH] KVM: VMX: enable LBR virtualization

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:	Thu, 15 Oct 2015 21:51:13 +0800
From:	Jian Zhou <jianjay.zhou@...wei.com>
To:	Paolo Bonzini <pbonzini@...hat.com>, <herongguang.he@...wei.com>,
	<zhang.zhanghailiang@...wei.com>, <gleb@...nel.org>,
	<tglx@...utronix.de>, <mingo@...hat.com>, <hpa@...or.com>,
	<x86@...nel.org>, <kvm@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>
CC:	<weidong.huang@...wei.com>, <peter.huangpeng@...wei.com>
Subject: Re: [PATCH] KVM: VMX: enable LBR virtualization

On 2015/10/14 19:30, Paolo Bonzini wrote:
>
>
> On 14/10/2015 13:26, Jian Zhou wrote:
>> On 12/10/2015 20:44, Paolo Bonzini wrote:
>>> In addition, the MSR numbers may differ between the guest and the host,
>>> because it is possible to emulate e.g. a Core CPU on a Core 2 CPU.  So I
>>> recommend against using the atomic switch mechanism for the from/to MSRs.
>>
>>    The vLBR feature depends on vPMU, and to enable vPMU, it needs to
>>    specify the "cpu mode" in the guest XML as host-passthrough. I think
>>    the MSR numbers between the guest and the host are the same in this
>>    senario.
>
> Does it depend on vPMU _for Linux guests_ or in general?  My impression
> is that LBR can be used by the guest independent of the PMU.

   I think only for Linux guests.

   I googled how to enable LBR on other guests(except Linux guests),
   e.g. Windows, and got no developer manuals about it.

   Here is an article about it:
   http://www.codeproject.com/Articles/517466/Last-branch-records-
   and-branch-tracing
   it says:
   "bit 8 of DR7 represents bit 0 of DebugCtl. This is the LBR bit."

   Intel developer manual vol 3B introduced DR7(Debug Control Register)
   and bit 8 of it on Section 17.2.4:
   "LE and GE (local and global exact breakpoint enable) flags (bits 8,
   9) — When set, these flags cause the processor to detect the exact
   instruction that caused a data breakpoint condition. For backward and
   forward compatibility with other Intel processors, we recommend that
   the LE and GE flags be set to 1 if exact breakpoints are required."

   But for now, I don't know how to test bit 8 of DR7 on Windows.

   Regards,

   Jian

>
>
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/