[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87pmu1ivvj.fsf@ashishki-desk.ger.corp.intel.com>
Date: Wed, 25 Aug 2021 14:53:04 +0300
From: Alexander Shishkin <alexander.shishkin@...ux.intel.com>
To: Xiaoyao Li <xiaoyao.li@...el.com>,
Like Xu <like.xu.linux@...il.com>
Cc: Sean Christopherson <seanjc@...gle.com>,
Vitaly Kuznetsov <vkuznets@...hat.com>,
Wanpeng Li <wanpengli@...cent.com>,
Jim Mattson <jmattson@...gle.com>,
Joerg Roedel <joro@...tes.org>, kvm@...r.kernel.org,
linux-kernel@...r.kernel.org, Paolo Bonzini <pbonzini@...hat.com>,
alexander.shishkin@...ux.intel.com
Subject: Re: [PATCH 3/5] KVM: VMX: RTIT_CTL_BRANCH_EN has no dependency on
other CPUID bit
Xiaoyao Li <xiaoyao.li@...el.com> writes:
> On 8/25/2021 2:08 PM, Like Xu wrote:
>> On 25/8/2021 12:19 pm, Xiaoyao Li wrote:
>>> On 8/25/2021 11:30 AM, Like Xu wrote:
>>> BranchEn should be always supported if PT is available. Per "31.2.7.2
>>
>> Check d35869ba348d3f1ff3e6d8214fe0f674bb0e404e.
>
> This commit shows BranchEn is supported on BDW, and must be enabled on
> BDW. This doesn't conflict the description above that BranchEn should be
> always supported.
It's the *not* setting BranchEn that's not supported on BDW. The point
of BranchEn is to allow the user to not set it and filter out all the
branch trace related packets. The main point of PT, however, is the
branch trace, so in the first implementation BranchEn was reserved as
1.
IOW, it's always available, doesn't depend on CPUID, but on BDW,
BranchEn==0 should throw a #GP, if I remember right. Check BDM106 for
details.
Regards,
--
Alex
Powered by blists - more mailing lists