Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.32 as permitted sender) client-ip=23.128.96.32;
From:   Tony Luck <tony.luck@intel.com>
To:     Fenghua Yu <fenghua.yu@intel.com>,
        Reinette Chatre <reinette.chatre@intel.com>,
        Peter Newman <peternewman@google.com>,
        Jonathan Corbet <corbet@lwn.net>,
        Shuah Khan <skhan@linuxfoundation.org>, x86@kernel.org
Cc:     Shaopeng Tan <tan.shaopeng@fujitsu.com>,
        James Morse <james.morse@arm.com>,
        Jamie Iles <quic_jiles@quicinc.com>,
        Babu Moger <babu.moger@amd.com>,
        Randy Dunlap <rdunlap@infradead.org>,
        linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
        patches@lists.linux.dev, Tony Luck <tony.luck@intel.com>
Subject: [PATCH v6 0/3] x86/resctrl: mba_MBps enhancements
Date:   Thu,  7 Dec 2023 11:56:10 -0800
Message-ID: <20231207195613.153980-1-tony.luck@intel.com>
In-Reply-To: <20231201214737.104444-1-tony.luck@intel.com>
References: <20231201214737.104444-1-tony.luck@intel.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Two changes relating to the MBA Software Controller(mba_sc):

1) Add a new mount option so the user can choose which memory
bandwidth monitoring event to use as the input to the feedback
loop.

2) Update the "mba_MBps" mount option to make use of total memory
bandwidth event on systems that do not support local bandwidth
event.

Signed-off-by: Tony Luck <tony.luck@intel.com>

---
Changes since v5:

Babu: Split into separate patches for the new mount option and the
update to the exising mount option. Since this is now a series, I
also moved the Documentation change to its own patch.

Reinette: Use "mbm_local_bytes", "mbm_total_bytes" as the strings for
the new "mba_MBps_event" option. This sets a precedent that new events
should follow the naming convention for the monitor files for the event.

Reinette: Update rdtgroup_show_options(). I'd completely missed this in
the earlier versions. Note that when the legacy "mba_MBps" mount option
is used, this will show in /proc as if the new option was used:

	# mount -t resctrl -o mba_MBps resctrl /sys/fs/resctrl/
	# grep resctrl /proc/mounts
	resctrl /sys/fs/resctrl resctrl rw,relatime,mba_MBps_event=mbm_local_bytes 0 0

Changing this to exacly match what the user typed would lose the detail
of which event is being used.

Reinette: More documentation needed. I added some text on why using
total instead of local might be useful to some users.

Tony Luck (3):
  x86/resctrl: Add mount option "mba_MBps_event"
  x86/resctrl: Use total bandwidth for mba_MBps option when local isn't
    present
  x86/resctrl: Add new "mba_MBps_event" mount option to documentation

 Documentation/arch/x86/resctrl.rst     | 16 ++++++-
 include/linux/resctrl.h                |  2 +
 arch/x86/kernel/cpu/resctrl/internal.h |  3 +-
 arch/x86/kernel/cpu/resctrl/monitor.c  | 21 ++++-----
 arch/x86/kernel/cpu/resctrl/rdtgroup.c | 63 +++++++++++++++++++++-----
 5 files changed, 79 insertions(+), 26 deletions(-)


base-commit: 33cc938e65a98f1d29d0a18403dbbee050dcad9a
-- 
2.41.0