From: Ratheesh Kannoth <rkannoth@marvell.com>
To: <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Cc: <sgoutham@marvell.com>, <davem@davemloft.net>,
<edumazet@google.com>, <kuba@kernel.org>, <pabeni@redhat.com>,
<horms@kernel.org>, <donald.hunter@gmail.com>, <jiri@resnulli.us>,
<chuck.lever@oracle.com>, <matttbe@kernel.org>,
<cjubran@nvidia.com>, <shshitrit@nvidia.com>,
<dtatulea@nvidia.com>, <tariqt@nvidia.com>,
Ratheesh Kannoth <rkannoth@marvell.com>
Subject: [PATCH v8 net-next 0/6] octeontx2-af: npc: Enhancements.
Date: Wed, 25 Mar 2026 12:51:53 +0530 [thread overview]
Message-ID: <20260325072159.1126964-1-rkannoth@marvell.com> (raw)
This series extends Marvell octeontx2-af support for CN20K NPC (MCAM
debuggability, allocation policy, default-rule lifetime, and optional
KPU profiles from firmware files) and adds a devlink mechanism for
multi-value parameters, with a small mlx5 follow-up to keep stack usage
within -Wframe-larger-than limits once union devlink_param_value grows.
Patch 1 improves CN20K MCAM visibility in debugfs: mcam_layout marks
enabled entries, dstats reports per-entry hit deltas, and mismatch lists
enabled entries without a PF mapping. MCAM enable state is tracked in a
bitmap updated from the CN20K enable path.
Patch 2 heap-allocates the temporary devlink param value array in
mlx5e_pcie_cong_get_thresh_config() so a larger union devlink_param_value
does not overflow the stack (patch 3).
Patch 3 (Saeed) introduces DEVLINK_PARAM_TYPE_U64_ARRAY and nested
DEVLINK_ATTR_PARAM_VALUE_DATA attributes so drivers and user space can
exchange bounded u64 arrays; devlink_nl_param_fill() moves large locals
to the heap as well.
Patch 4 adds a runtime devlink parameter npc_srch_order (U64 array) to
reorder CN20K subbank search during MCAM allocation.
Patch 5 ties default MCAM entries (broadcast, multicast, promisc, ucast)
to NIX LF alloc/free on CN20K, adds NIX_LF_DONT_FREE_DFT_IDXS for kernel
PF suspend-style teardown, and adjusts free-all and default-entry paths
so default rules are not freed as ordinary user rules.
Patch 6 allows loading a custom KPU profile from /lib/firmware/kpu via
module parameter kpu_profile on non-CN20K paths, with cam2/ptype support
and shared helpers for firmware-sourced vs filesystem-sourced profiles;
CN20K continues to use its existing custom KPU apply path.
The mlx5 change is placed immediately before the devlink union growth
so the series applies cleanly and stays warning-free when built
incrementally.
Ratheesh Kannoth (5):
octeontx2-af: npc: cn20k: debugfs enhancements
net/mlx5e: heap-allocate devlink param values
octeontx2-af: npc: cn20k: add subbank search order control
octeontx2-af: npc: cn20k: dynamically allocate and free default MCAM
entries
octeontx2-af: npc: Support for custom KPU profile from filesystem
Saeed Mahameed (1):
devlink: Implement devlink param multi attribute nested data values
Documentation/netlink/specs/devlink.yaml | 4 +
.../marvell/octeontx2/af/cn20k/debugfs.c | 126 ++++-
.../ethernet/marvell/octeontx2/af/cn20k/npc.c | 255 +++++++--
.../ethernet/marvell/octeontx2/af/cn20k/npc.h | 10 +
.../net/ethernet/marvell/octeontx2/af/mbox.h | 1 +
.../net/ethernet/marvell/octeontx2/af/npc.h | 17 +
.../net/ethernet/marvell/octeontx2/af/rvu.h | 6 +-
.../marvell/octeontx2/af/rvu_devlink.c | 92 +++-
.../ethernet/marvell/octeontx2/af/rvu_nix.c | 16 +-
.../ethernet/marvell/octeontx2/af/rvu_npc.c | 521 ++++++++++++++----
.../ethernet/marvell/octeontx2/af/rvu_npc.h | 17 +
.../ethernet/marvell/octeontx2/af/rvu_reg.h | 1 +
.../ethernet/marvell/octeontx2/nic/otx2_pf.c | 6 +-
.../mellanox/mlx5/core/en/pcie_cong_event.c | 11 +-
include/net/devlink.h | 8 +
include/uapi/linux/devlink.h | 1 +
net/devlink/netlink_gen.c | 2 +
net/devlink/param.c | 91 ++-
18 files changed, 979 insertions(+), 206 deletions(-)
--
v7 -> v8: Addressed Simon comments
https://lore.kernel.org/netdev/20260323035110.3908741-1-rkannoth@marvell.com/T/#t
v6 -> v7: Addressed Simon comments
https://lore.kernel.org/netdev/20260320165432.98832-1-horms@kernel.org/
v5 -> v6: Addressed Jakub,Jiri comments
https://lore.kernel.org/netdev/20260317045623.250187-1-rkannoth@marvell.com/
v4 -> v5: Addressed Jakub comments
https://lore.kernel.org/netdev/20260312022754.2029595-6-rkannoth@marvell.com/
v3 -> v4: Addressed Simon comments
https://lore.kernel.org/netdev/abDeXLpMMxp7G1v3@rkannoth-OptiPlex-7090/#t
v2 -> v3: Addressed Simon comments.
https://lore.kernel.org/netdev/20260304043032.3661647-1-rkannoth@marvell.com/
v1 -> v2: Addressed Jakub comments.
https://lore.kernel.org/netdev/20260302085803.2449828-1-rkannoth@marvell.com/#t
2.43.0
next reply other threads:[~2026-03-25 7:22 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-25 7:21 Ratheesh Kannoth [this message]
2026-03-25 7:21 ` [PATCH v8 net-next 1/6] octeontx2-af: npc: cn20k: debugfs enhancements Ratheesh Kannoth
2026-03-25 7:21 ` [PATCH v8 net-next 2/6] net/mlx5e: heap-allocate devlink param values Ratheesh Kannoth
2026-03-25 7:21 ` [PATCH v8 net-next 3/6] devlink: Implement devlink param multi attribute nested data values Ratheesh Kannoth
2026-03-25 7:21 ` [PATCH v8 net-next 4/6] octeontx2-af: npc: cn20k: add subbank search order control Ratheesh Kannoth
2026-03-25 7:21 ` [PATCH v8 net-next 5/6] octeontx2-af: npc: cn20k: dynamically allocate and free default MCAM entries Ratheesh Kannoth
2026-03-25 7:21 ` [PATCH v8 net-next 6/6] octeontx2-af: npc: Support for custom KPU profile from filesystem Ratheesh Kannoth
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=20260325072159.1126964-1-rkannoth@marvell.com \
--to=rkannoth@marvell.com \
--cc=chuck.lever@oracle.com \
--cc=cjubran@nvidia.com \
--cc=davem@davemloft.net \
--cc=donald.hunter@gmail.com \
--cc=dtatulea@nvidia.com \
--cc=edumazet@google.com \
--cc=horms@kernel.org \
--cc=jiri@resnulli.us \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=matttbe@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=sgoutham@marvell.com \
--cc=shshitrit@nvidia.com \
--cc=tariqt@nvidia.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