[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <4ada5bf1-87ab-4beb-8d25-f3c93c8fe4d3@amd.com>
Date: Fri, 18 Oct 2024 08:46:36 -0500
From: Mario Limonciello <mario.limonciello@....com>
To: Ricardo Neri <ricardo.neri-calderon@...ux.intel.com>
Cc: Borislav Petkov <bp@...en8.de>, Hans de Goede <hdegoede@...hat.com>,
Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>,
x86@...nel.org, "Gautham R . Shenoy" <gautham.shenoy@....com>,
Perry Yuan <perry.yuan@....com>, linux-kernel@...r.kernel.org,
linux-doc@...r.kernel.org, linux-pm@...r.kernel.org,
platform-driver-x86@...r.kernel.org,
Shyam Sundar S K <Shyam-sundar.S-k@....com>
Subject: Re: [PATCH v2 05/13] platform/x86: hfi: Introduce AMD Hardware
Feedback Interface Driver
On 10/17/2024 18:33, Ricardo Neri wrote:
> On Tue, Oct 15, 2024 at 01:09:42PM -0500, Mario Limonciello wrote:
>>>
>>>> Link: https://bugzilla.kernel.org/show_bug.cgi?id=206537
>>>
>>> I tried to find the HFI details on the documents in this "bug" but I could
>>> not find them. What document in specific could I look at?
>>>
>>> Thanks and BR,
>>> Ricardo
>>
>> Hi Ricardo,
>>
>> It is spread out across multiple places. This is part of the reason for
>> patch 1 in the series outlines details of how it works.
>>
>> The reason for that "collect all" Bugzilla for documentation is because the
>> URLs for AMD documentation have undergone changes in the past and it makes
>> it difficult to put stable URLs in commit messages. So teams that want to
>> reference documentation put it on a dump all bug for a stable URL to
>> reference.
>>
>> On that link you will find the APM, which will have some documentation
>> specifically for the CPUID leafs used for topology identification and
>> clearing history.
>>
>> Read patch 1 and let me know if it covers what specifically you're looking
>> for. If it's still missing some info let me know what you would like added.
>
> Thank you for your reply! I read patch 1. I was wondering specifically about
> more details of the Class ID. I see that they have associated counters and
> desired scheduling behavior.
Ah thanks! Obviously in this version there is no utilization of the
classifications, so this was an oversight. This is something that we'll
worry about after the baseline support is landed.
I'll make sure the documentation is updated in the next revision to
explain these.
>
> I was also curious about the layout of the HFI table. I guess I can infer it
> from patches 5 and 6 but if there is a picture already, I wouldn't mind. ;)
>
There's no picture right now, but in v3 I added a patch at the end of
the series to dump the table. I'll see what makes sense to add to
documentation.
Powered by blists - more mailing lists