mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Tony Luck <tony.luck@intel.com>
To: Fenghua Yu <fenghuay@nvidia.com>,
	Reinette Chatre <reinette.chatre@intel.com>,
	Maciej Wieczor-Retman <maciej.wieczor-retman@intel.com>,
	Peter Newman <peternewman@google.com>,
	James Morse <james.morse@arm.com>,
	Babu Moger <babu.moger@amd.com>,
	Drew Fustini <dfustini@baylibre.com>,
	Dave Martin <Dave.Martin@arm.com>, Chen Yu <yu.c.chen@intel.com>
Cc: x86@kernel.org, linux-kernel@vger.kernel.org,
	patches@lists.linux.dev, Tony Luck <tony.luck@intel.com>
Subject: [RFC PATCH v2 00/11] x86/resctrl: Support asymmetric L2 CAT on hybrid CPUs
Date: Tue,  3 Nov 2026 13:22:43 -0800	[thread overview]
Message-ID: <20261103212254.18314-1-tony.luck@intel.com> (raw)

Stil RFC as it builds on Reinette's multiple control patches[0].

Some hybrid CPUs (e.g. Panther Lake) have L2 caches whose cache
allocation capabilities differ between P-core and E-core domains.
Today these show up as symmetric L2 CAT, on just the P-core. New CPUID
leaves enumerate asymmetric allocation separately, including per-domain
capacity bitmask (CBM) lengths.

This series teaches resctrl to use that enumeration:

 - Define the four Intel RDT CPUID leaves, and add the asymmetric
   L2 CAT/CDP feature bits, CPUID dependencies and rdt= boot options
   (l2cat_asym, l2cdp_asym).
 - Disable symmetric L2 CAT on Panther Lake, where it is misreported.
 - Clean up rdt_get_cache_alloc_cfg() (failure handling, leaf argument)
   and the RDT allocation feature checks to prepare for a second leaf.
 - Move the CBM length from the shared control to each control domain
   (touches fs/resctrl, x86 and MPAM) since lengths can now differ.
 - Enumerate asymmetric L2 CAT/CDP from CPUID leaf 0x28 when no
   shareable bits are reported (resctrl cannot represent them per
   domain). info/L2/cbm_mask reports the minimum CBM across domains.
 - Change the "max" file to show per-domain values, matching the
   schemata format, so the UI needs no change for asymmetric controls.
 - Document the new options and the cbm_mask semantics.

Changes since first RFC[1]:

*) No new/renamed resource or control. Just keep this as "L2" for both.

*) Keep the info/L2/cbm_mask but make it display the minimum mask across
   all domains. Naive legacy s/w may use this and waste some cache.

*) Chen Yu pointed out that I don't need to search the resource for the
   control to update the domain. Drop the helper and just update the
   bitmask length in domain_add_cpu_ctrl()

*) Find per-domain cbm_len for L3 as well as L2 cache (though still only
   look for asymmetric L2

*) Intel L2 caches don't have any shareable bits. Check that the
   shareable mask is zero (as this series doesn't update file system
   code to handle asymmetric shareable bits).

[0] Link: git://git.kernel.org/pub/scm/linux/kernel/git/reinette/linux.git resctrl/controls_rfc_v2.4
[1] Link: https://lore.kernel.org/all/auDiLEujZ5BGCACj@agluck-desk3/

Tony Luck (11):
  fs/resctrl: Change UI of "max" file (merge back into earlier patches)
  x86/resctrl: Disable symmetric L2 CAT on Panther Lake
  x86/cpuid: Define Intel RDT enumeration leaves
  x86/resctrl: Add asymmetric L2 CAT feature bits and boot options
  x86/resctrl: Handle cache allocation setup failures
  x86/resctrl: Pass the CPUID leaf to rdt_get_cache_alloc_cfg()
  arm,x86,fs/resctrl: Store cache bitmask length per control domain
  x86/resctrl: Use cpu_feature_enabled() for RDT allocation
  x86/resctrl: Accept asymmetric RDT allocation features
  x86/resctrl: Enable asymmetric CAT from CPUID leaf 0x28
  fs/resctrl: Document asymmetric L2 allocation options

 .../admin-guide/kernel-parameters.txt         |   3 +-
 Documentation/filesystems/resctrl.rst         |   8 +-
 include/linux/resctrl.h                       |  12 +-
 arch/x86/include/asm/cpufeatures.h            |   3 +
 arch/x86/include/asm/cpuid/types.h            |   4 +
 arch/x86/kernel/cpu/cpuid-deps.c              |   2 +
 arch/x86/kernel/cpu/resctrl/core.c            | 138 +++++++++++++-----
 arch/x86/kernel/cpu/resctrl/rdtgroup.c        |   2 +-
 arch/x86/kernel/cpu/scattered.c               |   3 +
 drivers/resctrl/mpam_resctrl.c                |  14 +-
 fs/resctrl/ctrlmondata.c                      |   9 +-
 fs/resctrl/pseudo_lock.c                      |   2 +-
 fs/resctrl/rdtgroup.c                         |  56 ++++---
 13 files changed, 184 insertions(+), 72 deletions(-)

-- 
2.56.0


             reply	other threads:[~2026-11-03 21:23 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-03 21:22 Tony Luck [this message]
2026-11-03 21:22 ` [RFC PATCH v2 01/11] fs/resctrl: Change UI of "max" file (merge back into earlier patches) Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 02/11] x86/resctrl: Disable symmetric L2 CAT on Panther Lake Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 03/11] x86/cpuid: Define Intel RDT enumeration leaves Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 04/11] x86/resctrl: Add asymmetric L2 CAT feature bits and boot options Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 05/11] x86/resctrl: Handle cache allocation setup failures Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 06/11] x86/resctrl: Pass the CPUID leaf to rdt_get_cache_alloc_cfg() Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 07/11] arm,x86,fs/resctrl: Store cache bitmask length per control domain Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 08/11] x86/resctrl: Use cpu_feature_enabled() for RDT allocation Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 09/11] x86/resctrl: Accept asymmetric RDT allocation features Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 10/11] x86/resctrl: Enable asymmetric CAT from CPUID leaf 0x28 Tony Luck
2026-11-03 21:22 ` [RFC PATCH v2 11/11] fs/resctrl: Document asymmetric L2 allocation options Tony Luck

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=20261103212254.18314-1-tony.luck@intel.com \
    --to=tony.luck@intel.com \
    --cc=Dave.Martin@arm.com \
    --cc=babu.moger@amd.com \
    --cc=dfustini@baylibre.com \
    --cc=fenghuay@nvidia.com \
    --cc=james.morse@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maciej.wieczor-retman@intel.com \
    --cc=patches@lists.linux.dev \
    --cc=peternewman@google.com \
    --cc=reinette.chatre@intel.com \
    --cc=x86@kernel.org \
    --cc=yu.c.chen@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

all inboxes | Powered by JetHome®