Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 19 Oct 2022 11:01:41 -0700
From:   "H. Peter Anvin" <hpa@zytor.com>
To:     "Li, Xin3" <xin3.li@intel.com>, "Gross, Jurgen" <jgross@suse.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "x86@kernel.org" <x86@kernel.org>
CC:     "tglx@linutronix.de" <tglx@linutronix.de>,
        "mingo@redhat.com" <mingo@redhat.com>,
        "bp@alien8.de" <bp@alien8.de>,
        "dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>,
        "peterz@infradead.org" <peterz@infradead.org>,
        "brgerst@gmail.com" <brgerst@gmail.com>,
        "Bae, Chang Seok" <chang.seok.bae@intel.com>
Subject: =?US-ASCII?Q?RE=3A_=5BPATCH_v4_5/5=5D_x86/gsseg=3A_use_the_LKGS_?= =?US-ASCII?Q?instruction_if_available_for_load=5Fgs=5Findex=28=29?=
User-Agent: K-9 Mail for Android
In-Reply-To: <BN6PR1101MB2161DD8F4F587C6FBFCA18EFA82B9@BN6PR1101MB2161.namprd11.prod.outlook.com>
References: <20221019095035.10823-1-xin3.li@intel.com> <20221019095035.10823-6-xin3.li@intel.com> <439a325f-6a00-845e-7850-43abfc60cd44@suse.com> <BN6PR1101MB2161DD8F4F587C6FBFCA18EFA82B9@BN6PR1101MB2161.namprd11.prod.outlook.com>
Message-ID: <FDC4F3E7-DEDD-41CC-93E6-8021F16B509D@zytor.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On October 19, 2022 10:45:07 AM PDT, "Li, Xin3" <xin3=2Eli@intel=2Ecom> wro=
te:
>> > +static inline void __init lkgs_init(void) { #ifdef
>> > +CONFIG_PARAVIRT_XXL #ifdef CONFIG_X86_64
>> > +	if (cpu_feature_enabled(X86_FEATURE_LKGS))
>> > +		pv_ops=2Ecpu=2Eload_gs_index =3D native_lkgs;
>>=20
>> For this to work correctly when running as a Xen PV guest, you need to =
add
>>=20
>> 	setup_clear_cpu_cap(X86_FEATURE_LKGS);
>>=20
>> to xen_init_capabilities() in arch/x86/xen/enlighten_pv=2Ec, as otherwi=
se the Xen
>> specific =2Eload_gs_index vector will be overwritten=2E
>
>Yeah, we definitely should add it to disable LKGS in a Xen PV guest=2E
>
>So does it mean that the Xen PV uses a black list during feature detectio=
n?
>If yes then new features are often required to be masked with an explicit
>call to setup_clear_cpu_cap=2E
>
>Wouldn't a white list be better?
>Then the job is more just on the Xen PV side, and it can selectively enab=
le
>a new feature, sometimes with Xen PV specific handling code added=2E
>
>Xin
>
>>=20
>>=20
>> Juergen
>

Most things don't frob the paravirt list=2E

Maybe we should make the paravirt frobbing a separate patch, at it is sepa=
rable=2E