[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <8be0d994-219a-467f-8303-cd8f6605b784@intel.com>
Date: Wed, 7 May 2025 20:31:15 -0700
From: Reinette Chatre <reinette.chatre@...el.com>
To: Tony Luck <tony.luck@...el.com>, Fenghua Yu <fenghuay@...dia.com>, "Maciej
Wieczor-Retman" <maciej.wieczor-retman@...el.com>, Peter Newman
<peternewman@...gle.com>, James Morse <james.morse@....com>, Babu Moger
<babu.moger@....com>, Drew Fustini <dfustini@...libre.com>, Dave Martin
<Dave.Martin@....com>, Anil Keshavamurthy <anil.s.keshavamurthy@...el.com>,
Chen Yu <yu.c.chen@...el.com>
CC: <x86@...nel.org>, <linux-kernel@...r.kernel.org>,
<patches@...ts.linux.dev>
Subject: Re: [PATCH v4 03/31] fs/resctrl: Clean up
rdtgroup_mba_mbps_event_{show,write}()
Hi Tony,
On 4/28/25 5:33 PM, Tony Luck wrote:
> These routines hard-code the two legacy mbm events.
No context, just a cryptic problem description.
>
> Change to allow for other mbm events in the future.
What is changed? How are other MBM events allowed in future?
Something to start with (incomplete and needs improvement still):
rdtgroup_mba_mbps_event_{show,write}() respectively shows and stores
which memory bandwidth event is used as input to the software feedback
loop that keeps memory bandwidth below the value specified in the
schemata file.
rdtgroup_mba_mbps_event_{show,write}() hard codes the two legacy MBM
events, MBM total bytes and MBM local bytes.
(Needs explanation why the hardcoding is a problem and how this is fixed.
Since this series is not adding any new MBM events it is unclear what
motivates this "in the future" fix as part of this telemetry work.)
Reinette
Powered by blists - more mailing lists