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;
MIME-Version: 1.0
In-Reply-To: <AT5PR8401MB1169FA8B55B36BAE8DF0F6EBAB710@AT5PR8401MB1169.NAMPRD84.PROD.OUTLOOK.COM>
References: <1461620099-11933-1-git-send-email-toshi.kani@hpe.com>
 <1461620099-11933-2-git-send-email-toshi.kani@hpe.com> <CAA9_cmenbY9GStibRjO0BfTq0kzzzqThiLeOygJsT9L03BFsBQ@mail.gmail.com>
 <1529350667.14039.119.camel@hpe.com> <CAPcyv4jbctFgCob58sjQZ8msGOT7GBe=+Z_2P6YvNrrK-ztMag@mail.gmail.com>
 <AT5PR8401MB1169FA8B55B36BAE8DF0F6EBAB710@AT5PR8401MB1169.NAMPRD84.PROD.OUTLOOK.COM>
From:   Dan Williams <dan.j.williams@intel.com>
Date:   Mon, 18 Jun 2018 14:46:58 -0700
Message-ID: <CAPcyv4goV+VR15qpio-wqDpDS3nXsfYEbTSk1Ws6DPp6ECKcRQ@mail.gmail.com>
Subject: Re: [PATCH v3 1/2] acpi/nfit: Update nfit driver to comply with ACPI 6.1
To:     "Elliott, Robert (Persistent Memory)" <elliott@hpe.com>
Cc:     "Kani, Toshi" <toshi.kani@hpe.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-nvdimm@lists.01.org" <linux-nvdimm@lists.01.org>,
        "Moore, Robert" <robert.moore@intel.com>,
        "Li, Juston" <juston.li@intel.com>,
        "rjw@rjwysocki.net" <rjw@rjwysocki.net>,
        "linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jun 18, 2018 at 2:37 PM, Elliott, Robert (Persistent Memory)
<elliott@hpe.com> wrote:
>
>
>> -----Original Message-----
>> From: Dan Williams [mailto:dan.j.williams@intel.com]
>> Sent: Monday, June 18, 2018 2:47 PM
>> To: Kani, Toshi <toshi.kani@hpe.com>
>> Cc: linux-kernel@vger.kernel.org; linux-nvdimm@lists.01.org; Moore, Robert
>> <robert.moore@intel.com>; Li, Juston <juston.li@intel.com>;
>> rjw@rjwysocki.net; linux-acpi@vger.kernel.org; Elliott, Robert (Persistent
>> Memory) <elliott@hpe.com>
>> Subject: Re: [PATCH v3 1/2] acpi/nfit: Update nfit driver to comply with
>> ACPI 6.1
>>
>> On Mon, Jun 18, 2018 at 12:39 PM, Kani, Toshi <toshi.kani@hpe.com> wrote:
>> > On Mon, 2018-06-18 at 12:01 -0700, Dan Williams wrote:
>> >> On Mon, Apr 25, 2016 at 2:43 PM Toshi Kani <toshi.kani@hpe.com> wrote:
>> >> >
>> >> > ACPI 6.1, Table 5-133, updates NVDIMM Control Region Structure
>> >> > as follows.
>> >> >  - Valid Fields, Manufacturing Location, and Manufacturing Date
>> >> >    are added from reserved range.  No change in the structure size.
>> >> >  - IDs (SPD values) are stored as arrays of bytes (i.e. big-endian
>> >> >    format).  The spec clarifies that they need to be represented
>> >> >    as arrays of bytes as well.
>> >> >
>> >>
>> >> Circling back on this a couple years too late... where are you reading
>> >> this "arrays of bytes" note. As far as I can see this is wrong. JEDEC
>> >> says that vendor id is stored LSB of the id is stored at the lowest
>> >> byte in SPD, which is little endian. So it seems Linux has showing the
>> >> incorrect value for a long time now.
>> >
>> > This follows ACPI 6.2a section 5.2.25.10 NVDIMM Representation Format,
>> > which Robert cited in his comment below:
>> > https://patchwork.kernel.org/patch/10237609/
>>
>> Right, the representation format has the fields big-endian for some
>> reason, but the individual values for sysfs should be show
>> little-endian as far as I can see. What am I missing?
>
> In practice, the serial numbers from three major DDR4 DIMM manufacturers
> are being assigned as big-endian, like in this set of NVDIMM-Ns:
> /sys/bus/nd/devices/nmem0/nfit/serial:0x122f8255
> /sys/bus/nd/devices/nmem1/nfit/serial:0x122f7f5e
> /sys/bus/nd/devices/nmem2/nfit/serial:0x122f818f
> /sys/bus/nd/devices/nmem3/nfit/serial:0x122f821c
> /sys/bus/nd/devices/nmem4/nfit/serial:0x122f817e
> /sys/bus/nd/devices/nmem5/nfit/serial:0x122f81cd
> /sys/bus/nd/devices/nmem6/nfit/serial:0x122f821e
> /sys/bus/nd/devices/nmem7/nfit/serial:0x122f819b
> /sys/bus/nd/devices/nmem8/nfit/serial:0x122f81a2
> /sys/bus/nd/devices/nmem9/nfit/serial:0x122f8198
> /sys/bus/nd/devices/nmem10/nfit/serial:0x122f8193
> /sys/bus/nd/devices/nmem11/nfit/serial:0x122f7f58
> /sys/bus/nd/devices/nmem12/nfit/serial:0x122f81cb
> /sys/bus/nd/devices/nmem13/nfit/serial:0x122f8181
> /sys/bus/nd/devices/nmem14/nfit/serial:0x122f8210
> /sys/bus/nd/devices/nmem15/nfit/serial:0x122f821f
>
> and this set of regular DIMMs:
> 396851B4
> 3968134C
> 396852DA
> 396850AB
> 39685A13
> 39685317
> 396852DD
> 396852D9

Let's take something simple like Vendor ID. What is the Vendor ID for
these DIMMs and what does Linux print in sysfs?

> Of the possible approaches for the sysfs nfit field decodes:
> fixed big-endian:
>   matches printed label content (text and barcode)
>   matches ACPI display advice for management tools
>   probably matches SMBIOS Serial Number string format (although
>     that depends on the system firmware)
>   requires user to know that this OS uses big-endian
>   has been upstream for a while now
> fixed little-endian:
>   harder to see that cd812f12 matches 122f81cd seen elsewhere
>   harder to notice that B4516839 is a peer of 4C136839
>   might match other little-endian-only OSes
>   requires user to know that this OS uses little-endian
> native endian:
>   most confusing
>   config/status files referencing the DIMMs are not portable
>   requires user to know that this OS uses native endianness
>   requires user to know the CPU endianness
>   was upstream several years ago

The time upstream is painful, but if Linux is needlessly swizzling the
bits this needs to be fixed.