From: Reinette Chatre <reinette.chatre@intel.com>
To: "Luck, Tony" <tony.luck@intel.com>,
Aaron Tomlin <atomlin@atomlin.com>, Babu Moger <bmoger@amd.com>
Cc: "Dave.Martin@arm.com" <Dave.Martin@arm.com>,
"james.morse@arm.com" <james.morse@arm.com>,
"babu.moger@amd.com" <babu.moger@amd.com>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"mingo@redhat.com" <mingo@redhat.com>,
"bp@alien8.de" <bp@alien8.de>,
"dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/2] x86/resctrl: Add io_alloc_min_cbm_all interface for CBM reset
Date: Mon, 10 Nov 2025 17:00:50 -0800 [thread overview]
Message-ID: <937022c8-82cb-4f4c-a2a3-ceaf99580cc6@intel.com> (raw)
In-Reply-To: <SJ1PR11MB6083F15A9FCB09AEC3A46827FCCFA@SJ1PR11MB6083.namprd11.prod.outlook.com>
On 11/10/25 4:51 PM, Luck, Tony wrote:
>> The goal is not just to "clear" the CBM, but to reset it to the smallest
>> valid mask.
>
> "smallest valid mask" seems like a fuzzy concept.
>
> Intel had one CPU that required 2 bits to be set in an L3 mask.
> Subsequent CPUs only requires 1 bit.
>
> AMD allows setting zero bits.
>
> So for AMD "smallest valid mask" is zero. Very simple there.
>
> But on Intel you must have either one or two bits set
> (but it doesn't matter which bits).
Right, and this is exposed to user space via, for example for L3,
/sys/fs/resctrl/info/L3/min_cbm_bits.
>
> In some cases you might want to pick a bit that
> overlaps with the space shared in the cache for
> non-CPU devices (typically top two valid bits of the
> mask, but should check CPUID leaf 0x10 for the exact
> bits).
... and this is exposed to user space via /sys/fs/resctrl/info/L3/shareable_bits
and for this new feature the domain specific values can be seen from
io_alloc_cbm.
>
> Or you might want a bit that does NOT overlap with those
> non-CPU allocations.
>
Seems like user space has needed information to perform a sane reset?
Reinette
next prev parent reply other threads:[~2025-11-11 1:00 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-07 1:23 [PATCH 0/2] x86/resctrl: Add io_alloc_min_cbm_all interface for CBM reset Aaron Tomlin
2025-11-07 1:24 ` [PATCH 1/2] " Aaron Tomlin
2025-11-07 22:02 ` Moger, Babu
2025-11-07 22:25 ` Aaron Tomlin
2025-11-07 23:08 ` Moger, Babu
2025-11-08 0:19 ` Aaron Tomlin
2025-11-08 0:42 ` Luck, Tony
2025-11-10 16:34 ` Babu Moger
2025-11-10 17:50 ` Luck, Tony
2025-11-10 19:42 ` Babu Moger
2025-11-10 19:56 ` Luck, Tony
2025-11-10 21:40 ` Reinette Chatre
2025-11-11 0:41 ` Aaron Tomlin
2025-11-11 0:51 ` Luck, Tony
2025-11-11 1:00 ` Reinette Chatre [this message]
2025-11-11 1:40 ` Luck, Tony
2025-11-11 2:37 ` Reinette Chatre
2025-11-11 2:44 ` Aaron Tomlin
2025-11-11 15:29 ` Babu Moger
2025-11-11 21:04 ` Babu Moger
2025-11-11 22:34 ` Aaron Tomlin
2025-11-11 23:21 ` Moger, Babu
2025-11-07 1:24 ` [PATCH 2/2] fs/resctrl: resctrl: Return -EINVAL for a missing seq_show implementation Aaron Tomlin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=937022c8-82cb-4f4c-a2a3-ceaf99580cc6@intel.com \
--to=reinette.chatre@intel.com \
--cc=Dave.Martin@arm.com \
--cc=atomlin@atomlin.com \
--cc=babu.moger@amd.com \
--cc=bmoger@amd.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=james.morse@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox