Message-ID: <595D0537.1080309@hpe.com>
Date: Wed, 05 Jul 2017 11:26:47 -0400
From: Linda Knippers <linda.knippers@hpe.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.6.0
MIME-Version: 1.0
To: Dan Williams <dan.j.williams@intel.com>,
        Jerry Hoemann <Jerry.Hoemann@hpe.com>
CC: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-nvdimm@lists.01.org" <linux-nvdimm@lists.01.org>
Subject: Re: [PATCH v4 3/6] libnvdimm, acpi, nfit: Add bus level dsm mask
 for pass thru.
References: <cover.1498810220.git.jerry.hoemann@hpe.com>
 <293ee143dcabb386ca06b384a384171c256a2ecc.1498810220.git.jerry.hoemann@hpe.com>
 <CAPcyv4gcpmmr9LdX3DBtdX_ncwoBmCdC7yVPpGdO6EP7N9NOKw@mail.gmail.com>
 <20170701195805.GA13259@anatevka.americas.hpqcorp.net>
 <CAPcyv4hAMHzi7OK-z-B9LpT83Q1evH2-B-NXUMB2ebtVbjOdsg@mail.gmail.com>
 <CAPcyv4hP0zuZ0bdcQ5R3K2p5CRfHKXMSC_ov1D4qGXsXAQen=A@mail.gmail.com>
 <20170701203813.GA13574@anatevka.americas.hpqcorp.net>
 <CAPcyv4iMNjea939-E8of0nuyze8nHMEOp3ecii8+Y3pyYE0Ecw@mail.gmail.com>
 <20170704200849.GA15713@anatevka.americas.hpqcorp.net>
 <CAPcyv4isR--WJVeVYngGnCGea8HbQzKozsDz4rn6XWGW5vv=OA@mail.gmail.com>
In-Reply-To: <CAPcyv4isR--WJVeVYngGnCGea8HbQzKozsDz4rn6XWGW5vv=OA@mail.gmail.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
SpamDiagnosticOutput: 1:99
SpamDiagnosticMetadata: NSPM
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 05 Jul 2017 15:27:14.6532 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DF4PR84MB0076
X-OriginatorOrg: hpe.com
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 5655
Lines: 122

On 07/04/2017 04:37 PM, Dan Williams wrote:
> On Tue, Jul 4, 2017 at 1:08 PM, Jerry Hoemann <jerry.hoemann@hpe.com> wrote:
>> On Sat, Jul 01, 2017 at 01:46:03PM -0700, Dan Williams wrote:
>>> On Sat, Jul 1, 2017 at 1:38 PM, Jerry Hoemann <jerry.hoemann@hpe.com> wrote:
>>>> On Sat, Jul 01, 2017 at 01:10:31PM -0700, Dan Williams wrote:
>>>>> On Sat, Jul 1, 2017 at 1:08 PM, Dan Williams <dan.j.williams@intel.com> wrote:
>>>>>> On Sat, Jul 1, 2017 at 12:58 PM, Jerry Hoemann <jerry.hoemann@hpe.com> wrote:
>>>>>>> On Fri, Jun 30, 2017 at 08:55:22PM -0700, Dan Williams wrote:
>>>>>>>
>>>>>>> ...
>>>>>>>
>>>>>>>> On Fri, Jun 30, 2017 at 9:09 AM, Jerry Hoemann <jerry.hoemann@hpe.com> wrote:
>>>>>>>>> +               if (cmd == ND_CMD_CALL)
>>>>>>>>> +                       dsm_mask = nd_desc->bus_dsm_mask;
>>>>>>>>>                 desc = nd_cmd_bus_desc(cmd);
>>>>>>>>>                 uuid = to_nfit_uuid(NFIT_DEV_BUS);
>>>>>>>>>                 handle = adev->handle;
>>>>>>>>> @@ -1613,6 +1615,7 @@ static void acpi_nfit_init_dsms(struct acpi_nfit_desc *acpi_desc)
>>>>>>>>>         struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc;
>>>>>>>>>         const u8 *uuid = to_nfit_uuid(NFIT_DEV_BUS);
>>>>>>>>>         struct acpi_device *adev;
>>>>>>>>> +       unsigned long dsm_mask;
>>>>>>>>>         int i;
>>>>>>>>>
>>>>>>>>>         nd_desc->cmd_mask = acpi_desc->bus_cmd_force_en;
>>>>>>>>> @@ -1624,6 +1627,11 @@ static void acpi_nfit_init_dsms(struct acpi_nfit_desc *acpi_desc)
>>>>>>>>>                 if (acpi_check_dsm(adev->handle, uuid, 1, 1ULL << i))
>>>>>>>>>                         set_bit(i, &nd_desc->cmd_mask);
>>>>>>>>>         set_bit(ND_CMD_CALL, &nd_desc->cmd_mask);
>>>>>>>>> +
>>>>>>>>> +       dsm_mask = 0x3bf;
>>>>>>>>
>>>>>>>> I went ahead and fixed this up to use dsm_mask defined like this:
>>>>>>>>
>>>>>>>> +       dsm_mask =
>>>>>>>> +               (1 << ND_CMD_ARS_CAP) |
>>>>>>>> +               (1 << ND_CMD_ARS_START) |
>>>>>>>> +               (1 << ND_CMD_ARS_STATUS) |
>>>>>>>> +               (1 << ND_CMD_CLEAR_ERROR) |
>>>>>>>> +               (1 << NFIT_CMD_TRANSLATE_SPA) |
>>>>>>>> +               (1 << NFIT_CMD_ARS_INJECT_SET) |
>>>>>>>> +               (1 << NFIT_CMD_ARS_INJECT_CLEAR) |
>>>>>>>> +               (1 << NFIT_CMD_ARS_INJECT_GET);
>>>>>>>>
>>>>>>>> This drops function number 0 which userspace has no need to call.
>>>>>>>
>>>>>>> Actually I like to call function 0.  Its an excellent test when
>>>>>>> modifying the code path as its a no side effects function whose output
>>>>>>> is known in advance and instantly recognizable.  I also use it when
>>>>>>> testing new firmware.
>>>>>>>
>>>>>>> What is the downside to allowing it?  What bad things happen?
>>>>>>
>>>>>> It allows implementations to bypass the standardization process and
>>>>>> ship new root DSMs. It's always possible to patch the kernel locally
>>>>>> for development, so I see no reason to ship this capability globally.
>>>>
>>>> I don't understand this comment, but I think your next comment
>>>> essentially says to disregard this comment?
>>>
>>> Yes, sorry.
>>>
>>>>> Actually, just the discovery portion does not lead to this leak, but
>>>>> it's redundant when we have the 'dsm_mask' sysfs attribute.
>>>>
>>>> No.  The generation of the mask in sysfs is not done by
>>>> executing the code in acpi_nfit_ctl.  One of the reasons I call
>>>> function 0 to test changes I am making to the ioctl path itself.
>>>> The sysfs has nothing to do with that path and cannot be used
>>>> to serve this purpose.
>>>>
>>>> And since the content of sysfs has been edited it also can not be
>>>> used as a basic test of firmware.
>>>>
>>>> What is the downside to allowing the calling of function 0?
>>>
>>> It needlessly expands the kernel ABI. I would suggest, if you want to
>>
>> No.  It is not needless.  It is not an ABI extension.
>> Same goes for the override feature.

I have never understood why allowing function 0 is considered harmful.
It is a standard function defined by ACPI in general and specifically
for NVDIMM Rood Device _DSMs.  It is also defined for each vendor-specific
DSM family.  It is not an ABI extension.  It is a standard.

> If the need is testing then we have a tools/testing/nvdimm for that.
> Of course it's an ABI extension, it allows userspace to discover DSM
> function numbers the kernel didn't know about at compile time.

It also allows user space to determine which DSMs are actually supported
by the platform, which may be a subset of the defined set, in a standard
way.  Exposing information only in /sys just makes it harder for people
writing software (tools, tests, whatever) that need to support more than
just Linux.

>> I hope that ACPI doesn't extend the specification in the future because
>> we'll just have to redo these patches yet again.
> 
> Hopefully this is the last ACPI spec version where we add new DSMs to
> the root device. 

I wouldn't bet on it.

> All future methods should be named methods like what
> the specification started doing for NVIDMM leaf devices with _LSI,
> _LSR, and _LSW.

Those methods started out as DSMs for a specific vendor and then became
standardized.  It would not surprise me if that's the path that is taken
as new NVDIMM technologies evolve and new functions may be required.  It's
not always clear on the outset what should be standardized.

Aggressively preventing extensibility, especially when it's actually part
of a standard, baffles me.

-- ljk
> _______________________________________________
> Linux-nvdimm mailing list
> Linux-nvdimm@lists.01.org
> https://lists.01.org/mailman/listinfo/linux-nvdimm
>