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;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (1.0)
Subject: Re: selftests/x86/fsgsbase_64 test problem
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <624b9e67-14ee-6882-e55e-f337ec2471d3@zytor.com>
Date:   Mon, 29 Jan 2018 10:26:44 -0800
Cc:     Andy Lutomirski <luto@kernel.org>, Borislav Petkov <bp@alien8.de>,
        Dan Rue <dan.rue@linaro.org>, Shuah Khan <shuah@kernel.org>,
        Ingo Molnar <mingo@kernel.org>,
        Dmitry Safonov <dsafonov@virtuozzo.com>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <F6FCC323-841E-4337-B1E7-8F2E502552D6@amacapital.net>
References: <20180126153631.ha7yc33fj5uhitjo@xps> <CALCETrXVFC=6wPNpL0Dc7pkSv4JSaoEZZcu3Rtw0sg7KJsE5Pg@mail.gmail.com> <CALCETrUi7Ub2TbFy3Cvj+j4VXZeYULPY+mgL7OX7bz9L8GO9ew@mail.gmail.com> <CALCETrWvkd68wPCxGwmzqgsLTr_59+=L9u8obwt+f+oUQwDY=w@mail.gmail.com> <CALCETrW5=K5wJqLRZACmJJ+LF8Nb+Yf3OJ1tTNwqiOCxHu3w1Q@mail.gmail.com> <CALCETrWq04AngUHkSv-bqDYXDSVH8N0bhXS_Nh63SpMmndexSA@mail.gmail.com> <CALCETrUcJ2dVkSXGhtRvQeYQZZhPca95VhjQB8j6dFN6hzceSQ@mail.gmail.com> <46328204-e363-e517-f30c-c8c94ac1442c@zytor.com> <CALCETrW6SY8ktqSA-r108WkNf15js4eu0grDds=JVbNtBdSGqA@mail.gmail.com> <624b9e67-14ee-6882-e55e-f337ec2471d3@zytor.com>
To:     "H. Peter Anvin" <hpa@zytor.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Jan 29, 2018, at 10:12 AM, H. Peter Anvin <hpa@zytor.com> wrote:
>=20
>> On 01/29/18 08:37, Andy Lutomirski wrote:
>>=20
>> That's what I thought, too, and the SDM does say that, but the SDM
>> says all kinds of not-quite-correct things about segmentation.
>>=20
>>> It is pretty much scratch space (I have
>>> suggested using it for the gsbase once all those issues get sorted out,
>>> because it lets the paranoid code do something like:
>>>=20
>>>        rdgsbase %rax
>>>        push %rax       /* Save old gsbase */
>>>        push %rax       /* Reserve space on stack */
>>>        sgdt -2(%rsp)   /* We don't care about the limit */
>>>        pop %rax        /* %rax <- gdtbase */
>>>        mov (%rax),%rax /* GDT[0] holds the gsbase for this cpu */
>>>        wrgsbase %rax
>>=20
>> That will utterly suck on non-UMIP machines that have
>> hypervisor-provided UMIP emulation.
>>=20
>=20
> Is that a valid thing to optimize for, especially given that paranoid
> entries aren't the most common anyway?
>=20

A bunch of people seem to care about NMI performance for perf.  And the curr=
ent patch set works without this trick.

FWIW, if we switch all entries to the entry text trampoline, we get direct p=
ercpu access for free.=