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;
Message-ID: <35961c2f-08ab-9036-bc0c-6ede269ad3ec@intel.com>
Date:   Tue, 18 Oct 2022 15:41:43 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.2.2
Subject: Re: [PATCHv10 09/15] x86: Expose untagging mask in
 /proc/$PID/arch_status
Content-Language: en-US
To:     "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc:     Dave Hansen <dave.hansen@linux.intel.com>,
        Andy Lutomirski <luto@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>, x86@kernel.org,
        Kostya Serebryany <kcc@google.com>,
        Andrey Ryabinin <ryabinin.a.a@gmail.com>,
        Andrey Konovalov <andreyknvl@gmail.com>,
        Alexander Potapenko <glider@google.com>,
        Taras Madan <tarasmadan@google.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        "H . J . Lu" <hjl.tools@gmail.com>,
        Andi Kleen <ak@linux.intel.com>,
        Rick Edgecombe <rick.p.edgecombe@intel.com>,
        Bharata B Rao <bharata@amd.com>,
        Jacob Pan <jacob.jun.pan@linux.intel.com>,
        Ashok Raj <ashok.raj@intel.com>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
References: <20221018113358.7833-1-kirill.shutemov@linux.intel.com>
 <20221018113358.7833-10-kirill.shutemov@linux.intel.com>
 <cd8939c7-6194-53f5-14e1-89bef945fb47@intel.com>
 <20221018222419.oy2z2wanvwp2pnrm@box.shutemov.name>
From:   Dave Hansen <dave.hansen@intel.com>
In-Reply-To: <20221018222419.oy2z2wanvwp2pnrm@box.shutemov.name>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 10/18/22 15:24, Kirill A. Shutemov wrote:
> On Tue, Oct 18, 2022 at 02:02:43PM -0700, Dave Hansen wrote:
>> On 10/18/22 04:33, Kirill A. Shutemov wrote:
>>> Add a line in /proc/$PID/arch_status to report untag_mask. It can be
>>> used to find out LAM status of the process from the outside. It is
>>> useful for debuggers.
>> Considering that address masking is not x86-specific, it seems like this
>> needs a better home (another file in /proc).
> In generic /proc/$PID/status?

Seems like a sane place to me.

> And I'm not sure if it is a good idea at this stage. Semantics around tags
> is not settled across architectures: somewhere it is per-thread, somewhere
> per-process, somewhere it is global.
> 
> Maybe keep it arch-specific?

Yeah, but all of those things could be served by a thread-specific ABI.
The per-thread ABI won't enumerate the scope of the thing, of course.
But, it _can_ precisely communicate what semantics the thread has.