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;
From:   "Bae, Chang Seok" <chang.seok.bae@intel.com>
To:     Andy Lutomirski <luto@kernel.org>
CC:     Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@kernel.org>,
        "H . Peter Anvin" <hpa@zytor.com>, Andi Kleen <ak@linux.intel.com>,
        "Metzger, Markus T" <markus.t.metzger@intel.com>,
        "Shankar, Ravi V" <ravi.v.shankar@intel.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Dave Hansen <dave.hansen@linux.intel.com>
Subject: Re: [PATCH v5 08/13] x86/fsgsbase/64: Introduce the
 FIND_PERCPU_BASE macro
Thread-Topic: [PATCH v5 08/13] x86/fsgsbase/64: Introduce the
 FIND_PERCPU_BASE macro
Thread-Index: AQHUunBnyK8xBeMjrkOJBZxNaNROKKXNJgUAgBFirgA=
Date:   Wed, 13 Feb 2019 18:46:59 +0000
Message-ID: <5474F9B1-6E8F-4677-976B-C10103175B05@intel.com>
References: <20190201205319.15995-1-chang.seok.bae@intel.com>
 <20190201205319.15995-10-chang.seok.bae@intel.com>
 <CALCETrU0LD_sZOM2d6dSbnvEbs=zsVxEiy6dPp0w+AJO2JZf3A@mail.gmail.com>
In-Reply-To: <CALCETrU0LD_sZOM2d6dSbnvEbs=zsVxEiy6dPp0w+AJO2JZf3A@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <53F8A4509FBB3D48B090E20927100E14@intel.com>
Content-Transfer-Encoding: 8BIT
MIME-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Feb 2, 2019, at 09:17, Andy Lutomirski <luto@kernel.org> wrote:
> 
> On Fri, Feb 1, 2019 at 12:55 PM Chang S. Bae <chang.seok.bae@intel.com> wrote:
>> 
>> +
>> +/*
>> + * CPU/node NR is loaded from the limit (size) field of a special segment
>> + * descriptor entry in GDT.
>> + */
>> +.macro LOAD_CPU_AND_NODE_SEG_LIMIT reg:req
>> +       movq    $__CPUNODE_SEG, \reg
>> +       lsl     \reg, \reg
>> +.endm
>> +
> 
> Please put the alternative in here instead of in FIND_PERCPU_BASE.

I would like to apply your comment though, build errors will come up due to the
__CPUNODE_SEG. So, still hoping the alternative line below is straightforward enough.
Will send out a revision, that reflects all other comments, shortly.

Chang

> 
>> +/*
>> + * Fetch the per-CPU GSBASE value for this processor and put it in @reg.
>> + * We normally use %gs for accessing per-CPU data, but we are setting up
>> + * %gs here and obviously can not use %gs itself to access per-CPU data.
>> + */
>> +.macro FIND_PERCPU_BASE reg:req
>> 
>> +       ALTERNATIVE \
>> +               "LOAD_CPU_AND_NODE_SEG_LIMIT \reg", \
>> +               "RDPID  \reg", \
>> +               X86_FEATURE_RDPID
>> +       andq    $VDSO_CPUNODE_MASK, \reg
>> +       movq    __per_cpu_offset(, \reg, 8), \reg
>> +.endm