[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <52df445f-c2cf-44da-aa6b-995a6beeffc6@intel.com>
Date: Mon, 4 Dec 2023 14:51:55 -0800
From: Reinette Chatre <reinette.chatre@...el.com>
To: Tony Luck <tony.luck@...el.com>
CC: "babu.moger@....com" <babu.moger@....com>,
"Yu, Fenghua" <fenghua.yu@...el.com>,
Peter Newman <peternewman@...gle.com>,
"Jonathan Corbet" <corbet@....net>,
Shuah Khan <skhan@...uxfoundation.org>,
"x86@...nel.org" <x86@...nel.org>,
Shaopeng Tan <tan.shaopeng@...itsu.com>,
James Morse <james.morse@....com>,
Jamie Iles <quic_jiles@...cinc.com>,
"Randy Dunlap" <rdunlap@...radead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>,
"patches@...ts.linux.dev" <patches@...ts.linux.dev>
Subject: Re: [PATCH v5] x86/resctrl: Add event choices for mba_MBps
On 12/4/2023 2:15 PM, Reinette Chatre wrote:
> On 12/4/2023 1:08 PM, Tony Luck wrote:
>>
>> B) "mba_MBps={local|total)" forces use of chosen event, fails if event
>> is unavailable.
>
> I assume you mean "mba_MBps_event={local|total}". This is my preference but
> I would like to learn more about Babu's "overkill" argument. I do believe that
> (B) is well motivated in [1] and has no impact on AMD.
> My uncertainty here (considering one motivation is to future proof it against
> adding more events) is the generic "local" and "total" names. Even so, all
> I could come up with are "local_bw" and "total_bw", which I do not think are
> improvements.
Thinking about this more the existing names in resctrl may help here
... "mbm_total_bytes" and "mbm_local_bytes"? Any new event would need to
integrate with these names.
Reinette
Powered by blists - more mailing lists