[PATCH iwl-next v2 0/5] ice: add support for devmem/io_uring Rx and Tx

[PATCH iwl-next v2 0/5] ice: add support for devmem/io_uring Rx and Tx

[Alexander Lobakin]

Now that ice uses libeth for managing Rx buffers and supports
configurable header split, it's ready to get support for sending
and receiving packets with unreadable (to the kernel) frags.

Extend libeth just a little bit to allow creating PPs with custom
memory providers and make sure ice works correctly with the netdev
ops locking. Then add the full set of queue_mgmt_ops and don't
unmap unreadable frags on Tx completion.
No perf regressions for the regular flows and no code duplication
implied.

Credits to the fbnic developers, which's code helped me understand
the memory providers and queue_mgmt_ops logics and served as
a reference.

Alexander Lobakin (5):
  libeth: pass Rx queue index to PP when creating a fill queue
  libeth: handle creating pools with unreadable buffers
  ice: migrate to netdev ops lock
  ice: implement Rx queue management ops
  ice: add support for transmitting unreadable frags

 drivers/net/ethernet/intel/ice/ice_lib.h    |  11 +-
 drivers/net/ethernet/intel/ice/ice_txrx.h   |   2 +
 drivers/net/ethernet/intel/idpf/idpf_txrx.h |   2 +
 include/net/libeth/rx.h                     |   2 +
 include/net/libeth/tx.h                     |   2 +-
 drivers/net/ethernet/intel/iavf/iavf_txrx.c |   1 +
 drivers/net/ethernet/intel/ice/ice_base.c   | 194 ++++++++++++++------
 drivers/net/ethernet/intel/ice/ice_lib.c    |  56 +++++-
 drivers/net/ethernet/intel/ice/ice_main.c   |  50 ++---
 drivers/net/ethernet/intel/ice/ice_sf_eth.c |   2 +
 drivers/net/ethernet/intel/ice/ice_txrx.c   |  43 +++--
 drivers/net/ethernet/intel/ice/ice_xsk.c    |   4 +-
 drivers/net/ethernet/intel/idpf/idpf_txrx.c |  13 ++
 drivers/net/ethernet/intel/libeth/rx.c      |  46 +++++
 14 files changed, 325 insertions(+), 103 deletions(-)

---
From v1[0]:
* rebase on top of the latest next-queue;
* fix a typo 'rxq_ixd' -> 'rxq_idx' (Tony).

Testing hints:
* regular Rx and Tx for regressions;
* <kernel root>/tools/testing/selftests/drivers/net/hw/ contains
  scripts for testing netmem Rx and Tx, namely devmem.py and
  iou-zcrx.py (read the documentation first).

[0] https://lore.kernel.org/intel-wired-lan/20251125173603.3834486-1-aleksander.lobakin@intel.com
-- 
2.52.0

[PATCH iwl-next v2 1/5] libeth: pass Rx queue index to PP when creating a fill queue

[Alexander Lobakin]

Since recently, page_pool_create() accepts optional stack index of
the Rx queue which the pool will be created for. It can then be
used on control path for stuff like memory providers.
Add the same field to libeth_fq and pass the index from all the
drivers using libeth for managing Rx to simplify implementing MP
support later.
idpf has one libeth_fq per buffer/fill queue and each Rx queue has
two fill queues, but since fill queues can never be shared, we can
store the corresponding Rx queue index there during the
initialization to pass it to libeth.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf_txrx.h |  2 ++
 include/net/libeth/rx.h                     |  2 ++
 drivers/net/ethernet/intel/iavf/iavf_txrx.c |  1 +
 drivers/net/ethernet/intel/ice/ice_base.c   |  2 ++
 drivers/net/ethernet/intel/idpf/idpf_txrx.c | 13 +++++++++++++
 drivers/net/ethernet/intel/libeth/rx.c      |  1 +
 6 files changed, 21 insertions(+)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 6796f010e382..0eaebac8ceae 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -748,6 +748,7 @@ libeth_cacheline_set_assert(struct idpf_tx_queue, 64,
  * @size: Length of descriptor ring in bytes
  * @dma: Physical address of ring
  * @q_vector: Backreference to associated vector
+ * @rxq_idx: stack index of the corresponding Rx queue
  * @rx_buffer_low_watermark: RX buffer low watermark
  * @rx_hbuf_size: Header buffer size
  * @rx_buf_size: Buffer size
@@ -791,6 +792,7 @@ struct idpf_buf_queue {
 	dma_addr_t dma;
 
 	struct idpf_q_vector *q_vector;
+	u16 rxq_idx;
 
 	u16 rx_buffer_low_watermark;
 	u16 rx_hbuf_size;
diff --git a/include/net/libeth/rx.h b/include/net/libeth/rx.h
index 0e736846c5e8..db838ef7f9bb 100644
--- a/include/net/libeth/rx.h
+++ b/include/net/libeth/rx.h
@@ -72,6 +72,7 @@ enum libeth_fqe_type {
  * @no_napi: the queue is not a data queue and does not have NAPI
  * @buf_len: HW-writeable length per each buffer
  * @nid: ID of the closest NUMA node with memory
+ * @idx: stack index of the corresponding Rx queue
  */
 struct libeth_fq {
 	struct_group_tagged(libeth_fq_fp, fp,
@@ -90,6 +91,7 @@ struct libeth_fq {
 
 	u32			buf_len;
 	int			nid;
+	u32			idx;
 };
 
 int libeth_rx_fq_create(struct libeth_fq *fq, void *napi_dev);
diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
index 275b11dd0c60..3d938d7ab2cc 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
@@ -771,6 +771,7 @@ int iavf_setup_rx_descriptors(struct iavf_ring *rx_ring)
 		.count		= rx_ring->count,
 		.buf_len	= LIBIE_MAX_RX_BUF_LEN,
 		.nid		= NUMA_NO_NODE,
+		.idx		= rx_ring->queue_index,
 	};
 	int ret;
 
diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index 6fb7051aa463..7097324c38f3 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -607,6 +607,7 @@ static int ice_rxq_pp_create(struct ice_rx_ring *rq)
 	struct libeth_fq fq = {
 		.count		= rq->count,
 		.nid		= NUMA_NO_NODE,
+		.idx		= rq->q_index,
 		.hsplit		= rq->vsi->hsplit,
 		.xdp		= ice_is_xdp_ena_vsi(rq->vsi),
 		.buf_len	= LIBIE_MAX_RX_BUF_LEN,
@@ -629,6 +630,7 @@ static int ice_rxq_pp_create(struct ice_rx_ring *rq)
 		.count		= rq->count,
 		.type		= LIBETH_FQE_HDR,
 		.nid		= NUMA_NO_NODE,
+		.idx		= rq->q_index,
 		.xdp		= ice_is_xdp_ena_vsi(rq->vsi),
 	};
 
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index 72215612b460..5dc41b7ba609 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -561,6 +561,7 @@ static int idpf_rx_hdr_buf_alloc_all(struct idpf_buf_queue *bufq)
 		.type	= LIBETH_FQE_HDR,
 		.xdp	= idpf_xdp_enabled(bufq->q_vector->vport),
 		.nid	= idpf_q_vector_to_mem(bufq->q_vector),
+		.idx	= bufq->rxq_idx,
 	};
 	int ret;
 
@@ -703,6 +704,7 @@ static int idpf_rx_bufs_init_singleq(struct idpf_rx_queue *rxq)
 		.type		= LIBETH_FQE_MTU,
 		.buf_len	= IDPF_RX_MAX_BUF_SZ,
 		.nid		= idpf_q_vector_to_mem(rxq->q_vector),
+		.idx		= rxq->idx,
 	};
 	int ret;
 
@@ -763,6 +765,7 @@ static int idpf_rx_bufs_init(struct idpf_buf_queue *bufq,
 		.hsplit		= idpf_queue_has(HSPLIT_EN, bufq),
 		.xdp		= idpf_xdp_enabled(bufq->q_vector->vport),
 		.nid		= idpf_q_vector_to_mem(bufq->q_vector),
+		.idx		= bufq->rxq_idx,
 	};
 	int ret;
 
@@ -1922,6 +1925,16 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport,
 							LIBETH_RX_LL_LEN;
 			idpf_rxq_set_descids(rsrc, q);
 		}
+
+		if (!idpf_is_queue_model_split(rsrc->rxq_model))
+			continue;
+
+		for (u32 j = 0; j < rsrc->num_bufqs_per_qgrp; j++) {
+			struct idpf_buf_queue *bufq;
+
+			bufq = &rx_qgrp->splitq.bufq_sets[j].bufq;
+			bufq->rxq_idx = rx_qgrp->splitq.rxq_sets[0]->rxq.idx;
+		}
 	}
 
 err_alloc:
diff --git a/drivers/net/ethernet/intel/libeth/rx.c b/drivers/net/ethernet/intel/libeth/rx.c
index 1d8248a31037..9ac3a1448b2f 100644
--- a/drivers/net/ethernet/intel/libeth/rx.c
+++ b/drivers/net/ethernet/intel/libeth/rx.c
@@ -157,6 +157,7 @@ int libeth_rx_fq_create(struct libeth_fq *fq, void *napi_dev)
 		.order		= LIBETH_RX_PAGE_ORDER,
 		.pool_size	= fq->count,
 		.nid		= fq->nid,
+		.queue_idx	= fq->idx,
 		.dev		= napi ? napi->dev->dev.parent : napi_dev,
 		.netdev		= napi ? napi->dev : NULL,
 		.napi		= napi,
-- 
2.52.0

RE: [Intel-wired-lan] [PATCH iwl-next v2 1/5] libeth: pass Rx queue index to PP when creating a fill queue

[Nowlin, Alexander]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Alexander Lobakin
> Sent: Thursday, December 4, 2025 7:51 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Kitszel, Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn <andrew+netdev@lunn.ch>; 
> David S. Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller, Jacob E 
> <jacob.e.keller@intel.com>; Loktionov, Aleksandr <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org; linux-
> kernel@vger.kernel.org
> Subject: [Intel-wired-lan] [PATCH iwl-next v2 1/5] libeth: pass Rx queue index to PP when creating a fill queue
> 
> Since recently, page_pool_create() accepts optional stack index of the Rx queue which the pool will be created for. It can then be used on control path for stuff like memory providers.
> Add the same field to libeth_fq and pass the index from all the drivers using libeth for managing Rx to simplify implementing MP support later.
> idpf has one libeth_fq per buffer/fill queue and each Rx queue has two fill queues, but since fill queues can never be shared, we can store the corresponding Rx queue index there during the initialization to pass > it to libeth.
> 
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>  drivers/net/ethernet/intel/idpf/idpf_txrx.h |  2 ++
>  include/net/libeth/rx.h                     |  2 ++
>  drivers/net/ethernet/intel/iavf/iavf_txrx.c |  1 +
>  drivers/net/ethernet/intel/ice/ice_base.c   |  2 ++
>  drivers/net/ethernet/intel/idpf/idpf_txrx.c | 13 +++++++++++++
>  drivers/net/ethernet/intel/libeth/rx.c      |  1 +
>  6 files changed, 21 insertions(+)

Tested-by: Alexander Nowlin <alexander.nowlin@intel.com>

[PATCH iwl-next v2 2/5] libeth: handle creating pools with unreadable buffers

[Alexander Lobakin]

libeth uses netmems for quite some time already, so in order to
support unreadable frags / memory providers, it only needs to set
PP_FLAG_ALLOW_UNREADABLE_NETMEM when needed.
Also add a couple sanity checks to make sure the driver didn't mess
up the configuration options and, in case when an MP is installed,
return the truesize always equal to PAGE_SIZE, so that
libeth_rx_alloc() will never try to allocate frags. Memory providers
manage buffers on their own and expect 1:1 buffer / HW Rx descriptor
association.

Bonus: mention in the libeth_sqe_type description that
LIBETH_SQE_EMPTY should also be used for netmem Tx SQEs -- they
don't need DMA unmapping.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
 include/net/libeth/tx.h                |  2 +-
 drivers/net/ethernet/intel/libeth/rx.c | 45 ++++++++++++++++++++++++++
 2 files changed, 46 insertions(+), 1 deletion(-)

diff --git a/include/net/libeth/tx.h b/include/net/libeth/tx.h
index c3db5c6f1641..a66fc2b3a114 100644
--- a/include/net/libeth/tx.h
+++ b/include/net/libeth/tx.h
@@ -12,7 +12,7 @@
 
 /**
  * enum libeth_sqe_type - type of &libeth_sqe to act on Tx completion
- * @LIBETH_SQE_EMPTY: unused/empty OR XDP_TX/XSk frame, no action required
+ * @LIBETH_SQE_EMPTY: empty OR netmem/XDP_TX/XSk frame, no action required
  * @LIBETH_SQE_CTX: context descriptor with empty SQE, no action required
  * @LIBETH_SQE_SLAB: kmalloc-allocated buffer, unmap and kfree()
  * @LIBETH_SQE_FRAG: mapped skb frag, only unmap DMA
diff --git a/drivers/net/ethernet/intel/libeth/rx.c b/drivers/net/ethernet/intel/libeth/rx.c
index 9ac3a1448b2f..9b45c9cdd599 100644
--- a/drivers/net/ethernet/intel/libeth/rx.c
+++ b/drivers/net/ethernet/intel/libeth/rx.c
@@ -6,6 +6,7 @@
 #include <linux/export.h>
 
 #include <net/libeth/rx.h>
+#include <net/netdev_queues.h>
 
 /* Rx buffer management */
 
@@ -139,9 +140,50 @@ static bool libeth_rx_page_pool_params_zc(struct libeth_fq *fq,
 	fq->buf_len = clamp(mtu, LIBETH_RX_BUF_STRIDE, max);
 	fq->truesize = fq->buf_len;
 
+	/*
+	 * Allow frags only for kernel pages. `fq->truesize == pp->max_len`
+	 * will always fall back to regular page_pool_alloc_netmems()
+	 * regardless of the MTU / FQ buffer size.
+	 */
+	if (pp->flags & PP_FLAG_ALLOW_UNREADABLE_NETMEM)
+		fq->truesize = pp->max_len;
+
 	return true;
 }
 
+/**
+ * libeth_rx_page_pool_check_unread - check input params for unreadable MPs
+ * @fq: buffer queue to check
+ * @pp: &page_pool_params for the queue
+ *
+ * Make sure we don't create an invalid pool with full-frame unreadable
+ * buffers, bidirectional unreadable buffers or so, and configure the
+ * ZC payload pool accordingly.
+ *
+ * Return: true on success, false on invalid input params.
+ */
+static bool libeth_rx_page_pool_check_unread(const struct libeth_fq *fq,
+					     struct page_pool_params *pp)
+{
+	if (!pp->netdev)
+		return true;
+
+	if (!netif_rxq_has_unreadable_mp(pp->netdev, pp->queue_idx))
+		return true;
+
+	/* For now, the core stack doesn't allow XDP with unreadable frags */
+	if (fq->xdp)
+		return false;
+
+	/* It should be either a header pool or a ZC payload pool */
+	if (fq->type == LIBETH_FQE_HDR)
+		return !fq->hsplit;
+
+	pp->flags |= PP_FLAG_ALLOW_UNREADABLE_NETMEM;
+
+	return fq->hsplit;
+}
+
 /**
  * libeth_rx_fq_create - create a PP with the default libeth settings
  * @fq: buffer queue struct to fill
@@ -166,6 +208,9 @@ int libeth_rx_fq_create(struct libeth_fq *fq, void *napi_dev)
 	struct page_pool *pool;
 	int ret;
 
+	if (!libeth_rx_page_pool_check_unread(fq, &pp))
+		return -EINVAL;
+
 	pp.dma_dir = fq->xdp ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE;
 
 	if (!fq->hsplit)
-- 
2.52.0

RE: [Intel-wired-lan] [PATCH iwl-next v2 2/5] libeth: handle creating pools with unreadable buffers

[Nowlin, Alexander]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Alexander Lobakin
> Sent: Thursday, December 4, 2025 7:52 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Kitszel, Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn <andrew+netdev@lunn.ch>; 
> David S. Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller, Jacob E 
> <jacob.e.keller@intel.com>; Loktionov, Aleksandr <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org; linux-
> kernel@vger.kernel.org
> Subject: [Intel-wired-lan] [PATCH iwl-next v2 2/5] libeth: handle creating pools with unreadable buffers
> 
> libeth uses netmems for quite some time already, so in order to support unreadable frags / memory providers, it only needs to set PP_FLAG_ALLOW_UNREADABLE_NETMEM when needed.
> Also add a couple sanity checks to make sure the driver didn't mess up the configuration options and, in case when an MP is installed, return the truesize always equal to PAGE_SIZE, so that
> libeth_rx_alloc() will never try to allocate frags. Memory providers manage buffers on their own and expect 1:1 buffer / HW Rx descriptor association.
> 
> Bonus: mention in the libeth_sqe_type description that LIBETH_SQE_EMPTY should also be used for netmem Tx SQEs -- they don't need DMA unmapping.
> 
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>  include/net/libeth/tx.h                |  2 +-
>  drivers/net/ethernet/intel/libeth/rx.c | 45 ++++++++++++++++++++++++++
>  2 files changed, 46 insertions(+), 1 deletion(-)

Tested-by: Alexander Nowlin <alexander.nowlin@intel.com>

[PATCH iwl-next v2 3/5] ice: migrate to netdev ops lock

[Alexander Lobakin]

Queue management ops unconditionally enable netdev locking. The same
lock is taken by default by several NAPI configuration functions,
such as napi_enable() and netif_napi_set_irq().
Request ops locking in advance and make sure we use the _locked
counterparts of those functions to avoid deadlocks, taking the lock
manually where needed (suspend/resume, queue rebuild and resets).

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_lib.h    |  6 ++-
 drivers/net/ethernet/intel/ice/ice_lib.c    | 56 +++++++++++++++++----
 drivers/net/ethernet/intel/ice/ice_main.c   | 49 ++++++++++--------
 drivers/net/ethernet/intel/ice/ice_sf_eth.c |  1 +
 drivers/net/ethernet/intel/ice/ice_xsk.c    |  4 +-
 5 files changed, 82 insertions(+), 34 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h
index 49454d98dcfe..347b63e497e7 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_lib.h
@@ -53,9 +53,11 @@ struct ice_vsi *
 ice_vsi_setup(struct ice_pf *pf, struct ice_vsi_cfg_params *params);
 
 void ice_vsi_set_napi_queues(struct ice_vsi *vsi);
-void ice_napi_add(struct ice_vsi *vsi);
-
+void ice_vsi_set_napi_queues_locked(struct ice_vsi *vsi);
 void ice_vsi_clear_napi_queues(struct ice_vsi *vsi);
+void ice_vsi_clear_napi_queues_locked(struct ice_vsi *vsi);
+
+void ice_napi_add(struct ice_vsi *vsi);
 
 int ice_vsi_release(struct ice_vsi *vsi);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 17d92ba65128..ac5d95a28f72 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2703,7 +2703,7 @@ void ice_vsi_close(struct ice_vsi *vsi)
 	if (!test_and_set_bit(ICE_VSI_DOWN, vsi->state))
 		ice_down(vsi);
 
-	ice_vsi_clear_napi_queues(vsi);
+	ice_vsi_clear_napi_queues_locked(vsi);
 	ice_vsi_free_irq(vsi);
 	ice_vsi_free_tx_rings(vsi);
 	ice_vsi_free_rx_rings(vsi);
@@ -2772,12 +2772,13 @@ void ice_dis_vsi(struct ice_vsi *vsi, bool locked)
 }
 
 /**
- * ice_vsi_set_napi_queues - associate netdev queues with napi
+ * ice_vsi_set_napi_queues_locked - associate netdev queues with napi
  * @vsi: VSI pointer
  *
  * Associate queue[s] with napi for all vectors.
+ * Must be called only with the netdev_lock taken.
  */
-void ice_vsi_set_napi_queues(struct ice_vsi *vsi)
+void ice_vsi_set_napi_queues_locked(struct ice_vsi *vsi)
 {
 	struct net_device *netdev = vsi->netdev;
 	int q_idx, v_idx;
@@ -2785,7 +2786,6 @@ void ice_vsi_set_napi_queues(struct ice_vsi *vsi)
 	if (!netdev)
 		return;
 
-	ASSERT_RTNL();
 	ice_for_each_rxq(vsi, q_idx)
 		netif_queue_set_napi(netdev, q_idx, NETDEV_QUEUE_TYPE_RX,
 				     &vsi->rx_rings[q_idx]->q_vector->napi);
@@ -2797,17 +2797,37 @@ void ice_vsi_set_napi_queues(struct ice_vsi *vsi)
 	ice_for_each_q_vector(vsi, v_idx) {
 		struct ice_q_vector *q_vector = vsi->q_vectors[v_idx];
 
-		netif_napi_set_irq(&q_vector->napi, q_vector->irq.virq);
+		netif_napi_set_irq_locked(&q_vector->napi, q_vector->irq.virq);
 	}
 }
 
 /**
- * ice_vsi_clear_napi_queues - dissociate netdev queues from napi
+ * ice_vsi_set_napi_queues - associate VSI queues with NAPIs
  * @vsi: VSI pointer
  *
+ * Version of ice_vsi_set_napi_queues_locked() that takes the netdev_lock,
+ * to use it outside of the net_device_ops context.
+ */
+void ice_vsi_set_napi_queues(struct ice_vsi *vsi)
+{
+	struct net_device *netdev = vsi->netdev;
+
+	if (!netdev)
+		return;
+
+	netdev_lock(netdev);
+	ice_vsi_set_napi_queues_locked(vsi);
+	netdev_unlock(netdev);
+}
+
+/**
+ * ice_vsi_clear_napi_queues_locked - dissociate netdev queues from napi
+ * @vsi: VSI to process
+ *
  * Clear the association between all VSI queues queue[s] and napi.
+ * Must be called only with the netdev_lock taken.
  */
-void ice_vsi_clear_napi_queues(struct ice_vsi *vsi)
+void ice_vsi_clear_napi_queues_locked(struct ice_vsi *vsi)
 {
 	struct net_device *netdev = vsi->netdev;
 	int q_idx, v_idx;
@@ -2815,12 +2835,11 @@ void ice_vsi_clear_napi_queues(struct ice_vsi *vsi)
 	if (!netdev)
 		return;
 
-	ASSERT_RTNL();
 	/* Clear the NAPI's interrupt number */
 	ice_for_each_q_vector(vsi, v_idx) {
 		struct ice_q_vector *q_vector = vsi->q_vectors[v_idx];
 
-		netif_napi_set_irq(&q_vector->napi, -1);
+		netif_napi_set_irq_locked(&q_vector->napi, -1);
 	}
 
 	ice_for_each_txq(vsi, q_idx)
@@ -2830,6 +2849,25 @@ void ice_vsi_clear_napi_queues(struct ice_vsi *vsi)
 		netif_queue_set_napi(netdev, q_idx, NETDEV_QUEUE_TYPE_RX, NULL);
 }
 
+/**
+ * ice_vsi_clear_napi_queues - dissociate VSI queues from NAPIs
+ * @vsi: VSI to process
+ *
+ * Version of ice_vsi_clear_napi_queues_locked() that takes the netdev lock,
+ * to use it outside of the net_device_ops context.
+ */
+void ice_vsi_clear_napi_queues(struct ice_vsi *vsi)
+{
+	struct net_device *netdev = vsi->netdev;
+
+	if (!netdev)
+		return;
+
+	netdev_lock(netdev);
+	ice_vsi_clear_napi_queues_locked(vsi);
+	netdev_unlock(netdev);
+}
+
 /**
  * ice_napi_add - register NAPI handler for the VSI
  * @vsi: VSI for which NAPI handler is to be registered
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 96730bf288fd..881b41d37821 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3523,6 +3523,7 @@ static void ice_set_ops(struct ice_vsi *vsi)
 	}
 
 	netdev->netdev_ops = &ice_netdev_ops;
+	netdev->request_ops_lock = true;
 	netdev->udp_tunnel_nic_info = &pf->hw.udp_tunnel_nic;
 	netdev->xdp_metadata_ops = &ice_xdp_md_ops;
 	ice_set_ethtool_ops(netdev);
@@ -5533,16 +5534,17 @@ static int ice_reinit_interrupt_scheme(struct ice_pf *pf)
 
 	/* Remap vectors and rings, after successful re-init interrupts */
 	ice_for_each_vsi(pf, v) {
-		if (!pf->vsi[v])
+		struct ice_vsi *vsi = pf->vsi[v];
+
+		if (!vsi)
 			continue;
 
-		ret = ice_vsi_alloc_q_vectors(pf->vsi[v]);
+		ret = ice_vsi_alloc_q_vectors(vsi);
 		if (ret)
 			goto err_reinit;
-		ice_vsi_map_rings_to_vectors(pf->vsi[v]);
-		rtnl_lock();
-		ice_vsi_set_napi_queues(pf->vsi[v]);
-		rtnl_unlock();
+
+		ice_vsi_map_rings_to_vectors(vsi);
+		ice_vsi_set_napi_queues(vsi);
 	}
 
 	ret = ice_req_irq_msix_misc(pf);
@@ -5555,13 +5557,15 @@ static int ice_reinit_interrupt_scheme(struct ice_pf *pf)
 	return 0;
 
 err_reinit:
-	while (v--)
-		if (pf->vsi[v]) {
-			rtnl_lock();
-			ice_vsi_clear_napi_queues(pf->vsi[v]);
-			rtnl_unlock();
-			ice_vsi_free_q_vectors(pf->vsi[v]);
-		}
+	while (v--) {
+		struct ice_vsi *vsi = pf->vsi[v];
+
+		if (!vsi)
+			continue;
+
+		ice_vsi_clear_napi_queues(vsi);
+		ice_vsi_free_q_vectors(vsi);
+	}
 
 	return ret;
 }
@@ -5623,14 +5627,17 @@ static int ice_suspend(struct device *dev)
 	 * to CPU0.
 	 */
 	ice_free_irq_msix_misc(pf);
+
 	ice_for_each_vsi(pf, v) {
-		if (!pf->vsi[v])
+		struct ice_vsi *vsi = pf->vsi[v];
+
+		if (!vsi)
 			continue;
-		rtnl_lock();
-		ice_vsi_clear_napi_queues(pf->vsi[v]);
-		rtnl_unlock();
-		ice_vsi_free_q_vectors(pf->vsi[v]);
+
+		ice_vsi_clear_napi_queues(vsi);
+		ice_vsi_free_q_vectors(vsi);
 	}
+
 	ice_clear_interrupt_scheme(pf);
 
 	pci_save_state(pdev);
@@ -6744,7 +6751,7 @@ static void ice_napi_enable_all(struct ice_vsi *vsi)
 		ice_init_moderation(q_vector);
 
 		if (q_vector->rx.rx_ring || q_vector->tx.tx_ring)
-			napi_enable(&q_vector->napi);
+			napi_enable_locked(&q_vector->napi);
 	}
 }
 
@@ -7243,7 +7250,7 @@ static void ice_napi_disable_all(struct ice_vsi *vsi)
 		struct ice_q_vector *q_vector = vsi->q_vectors[q_idx];
 
 		if (q_vector->rx.rx_ring || q_vector->tx.tx_ring)
-			napi_disable(&q_vector->napi);
+			napi_disable_locked(&q_vector->napi);
 
 		cancel_work_sync(&q_vector->tx.dim.work);
 		cancel_work_sync(&q_vector->rx.dim.work);
@@ -7543,7 +7550,7 @@ int ice_vsi_open(struct ice_vsi *vsi)
 		if (err)
 			goto err_set_qs;
 
-		ice_vsi_set_napi_queues(vsi);
+		ice_vsi_set_napi_queues_locked(vsi);
 	}
 
 	err = ice_up_complete(vsi);
diff --git a/drivers/net/ethernet/intel/ice/ice_sf_eth.c b/drivers/net/ethernet/intel/ice/ice_sf_eth.c
index 1a2c94375ca7..2c3db1b03055 100644
--- a/drivers/net/ethernet/intel/ice/ice_sf_eth.c
+++ b/drivers/net/ethernet/intel/ice/ice_sf_eth.c
@@ -58,6 +58,7 @@ static int ice_sf_cfg_netdev(struct ice_dynamic_port *dyn_port,
 	eth_hw_addr_set(netdev, dyn_port->hw_addr);
 	ether_addr_copy(netdev->perm_addr, dyn_port->hw_addr);
 	netdev->netdev_ops = &ice_sf_netdev_ops;
+	netdev->request_ops_lock = true;
 	SET_NETDEV_DEVLINK_PORT(netdev, devlink_port);
 
 	err = register_netdev(netdev);
diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c
index 953e68ed0f9a..6d08a11a86a8 100644
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@ -33,9 +33,9 @@ ice_qvec_toggle_napi(struct ice_vsi *vsi, struct ice_q_vector *q_vector,
 		return;
 
 	if (enable)
-		napi_enable(&q_vector->napi);
+		napi_enable_locked(&q_vector->napi);
 	else
-		napi_disable(&q_vector->napi);
+		napi_disable_locked(&q_vector->napi);
 }
 
 /**
-- 
2.52.0

Re: [Intel-wired-lan] [PATCH iwl-next v2 3/5] ice: migrate to netdev ops lock

[Alexander Lobakin]

From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu,  4 Dec 2025 16:51:31 +0100

> Queue management ops unconditionally enable netdev locking. The same
> lock is taken by default by several NAPI configuration functions,
> such as napi_enable() and netif_napi_set_irq().
> Request ops locking in advance and make sure we use the _locked
> counterparts of those functions to avoid deadlocks, taking the lock
> manually where needed (suspend/resume, queue rebuild and resets).
> 
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Note: Larysa found that this commit breaks `ethtool -L` -- the system
hangs. Seems like some of the functions called during the queue
reconfiguration still take the netdev lock and a deadlock happens
(I definitely tested `ethtool -G`, but might've forgotten to test
`-L`...).

I'll try to fix this ASAP and send a fixup patch. Since nobody (?)
reported this earlier, maybe it's not worth dropping the series from
the next-queue in the meantime...

Thanks,
Olek

[PATCH iwl-next] ice: fix system hang on `ethtool -L`

[Alexander Lobakin]

ice_set_channels() calls ice_vsi_rebuild() under the netdev lock
taken, but ice_vsi_rebuild() calls netif_napi_{add,del}() which
take the same lock.
Add ice_vsi_rebuild_locked() which uses the _locked counterparts
of these functions and use it in ice_set_channels().

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
Hey Tony, please amend to the patch I replied to.
---
 drivers/net/ethernet/intel/ice/ice_base.h |  2 +
 drivers/net/ethernet/intel/ice/ice_lib.h  |  1 +
 drivers/net/ethernet/intel/ice/ice_base.c | 63 ++++++++++++---
 drivers/net/ethernet/intel/ice/ice_lib.c  | 94 ++++++++++++++++++++---
 drivers/net/ethernet/intel/ice/ice_main.c |  5 +-
 5 files changed, 143 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_base.h b/drivers/net/ethernet/intel/ice/ice_base.h
index d28294247599..99b2c7232829 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.h
+++ b/drivers/net/ethernet/intel/ice/ice_base.h
@@ -12,8 +12,10 @@ int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg);
 int
 ice_vsi_ctrl_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx, bool wait);
 int ice_vsi_wait_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx);
+int ice_vsi_alloc_q_vectors_locked(struct ice_vsi *vsi);
 int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi);
 void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi);
+void ice_vsi_free_q_vectors_locked(struct ice_vsi *vsi);
 void ice_vsi_free_q_vectors(struct ice_vsi *vsi);
 int ice_vsi_cfg_single_txq(struct ice_vsi *vsi, struct ice_tx_ring **tx_rings,
 			   u16 q_idx);
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h
index 347b63e497e7..e55b72db72c4 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_lib.h
@@ -68,6 +68,7 @@ int ice_ena_vsi(struct ice_vsi *vsi, bool locked);
 void ice_vsi_decfg(struct ice_vsi *vsi);
 void ice_dis_vsi(struct ice_vsi *vsi, bool locked);
 
+int ice_vsi_rebuild_locked(struct ice_vsi *vsi, u32 vsi_flags);
 int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags);
 int ice_vsi_cfg(struct ice_vsi *vsi);
 struct ice_vsi *ice_vsi_alloc(struct ice_pf *pf);
diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index 7097324c38f3..65e19815bec5 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -153,8 +153,8 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, u16 v_idx)
 	 * handler here (i.e. resume, reset/rebuild, etc.)
 	 */
 	if (vsi->netdev)
-		netif_napi_add_config(vsi->netdev, &q_vector->napi,
-				      ice_napi_poll, v_idx);
+		netif_napi_add_config_locked(vsi->netdev, &q_vector->napi,
+					     ice_napi_poll, v_idx);
 
 out:
 	/* tie q_vector and VSI together */
@@ -196,7 +196,7 @@ static void ice_free_q_vector(struct ice_vsi *vsi, int v_idx)
 
 	/* only VSI with an associated netdev is set up with NAPI */
 	if (vsi->netdev)
-		netif_napi_del(&q_vector->napi);
+		netif_napi_del_locked(&q_vector->napi);
 
 	/* release MSIX interrupt if q_vector had interrupt allocated */
 	if (q_vector->irq.index < 0)
@@ -896,13 +896,15 @@ int ice_vsi_wait_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx)
 }
 
 /**
- * ice_vsi_alloc_q_vectors - Allocate memory for interrupt vectors
+ * ice_vsi_alloc_q_vectors_locked - Allocate memory for interrupt vectors
  * @vsi: the VSI being configured
  *
- * We allocate one q_vector per queue interrupt. If allocation fails we
- * return -ENOMEM.
+ * Should be called only under the netdev lock.
+ * We allocate one q_vector per queue interrupt.
+ *
+ * Return: 0 on success, -ENOMEM if allocation fails.
  */
-int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi)
+int ice_vsi_alloc_q_vectors_locked(struct ice_vsi *vsi)
 {
 	struct device *dev = ice_pf_to_dev(vsi->back);
 	u16 v_idx;
@@ -929,6 +931,30 @@ int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi)
 	return v_idx ? 0 : err;
 }
 
+/**
+ * ice_vsi_alloc_q_vectors - Allocate memory for interrupt vectors
+ * @vsi: the VSI being configured
+ *
+ * We allocate one q_vector per queue interrupt.
+ *
+ * Return: 0 on success, -ENOMEM if allocation fails.
+ */
+int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi)
+{
+	struct net_device *dev = vsi->netdev;
+	int ret;
+
+	if (dev)
+		netdev_lock(dev);
+
+	ret = ice_vsi_alloc_q_vectors_locked(vsi);
+
+	if (dev)
+		netdev_unlock(dev);
+
+	return ret;
+}
+
 /**
  * ice_vsi_map_rings_to_vectors - Map VSI rings to interrupt vectors
  * @vsi: the VSI being configured
@@ -992,10 +1018,12 @@ void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi)
 }
 
 /**
- * ice_vsi_free_q_vectors - Free memory allocated for interrupt vectors
+ * ice_vsi_free_q_vectors_locked - Free memory allocated for interrupt vectors
  * @vsi: the VSI having memory freed
+ *
+ * Should be called only under the netdev lock.
  */
-void ice_vsi_free_q_vectors(struct ice_vsi *vsi)
+void ice_vsi_free_q_vectors_locked(struct ice_vsi *vsi)
 {
 	int v_idx;
 
@@ -1005,6 +1033,23 @@ void ice_vsi_free_q_vectors(struct ice_vsi *vsi)
 	vsi->num_q_vectors = 0;
 }
 
+/**
+ * ice_vsi_free_q_vectors - Free memory allocated for interrupt vectors
+ * @vsi: the VSI having memory freed
+ */
+void ice_vsi_free_q_vectors(struct ice_vsi *vsi)
+{
+	struct net_device *dev = vsi->netdev;
+
+	if (dev)
+		netdev_lock(dev);
+
+	ice_vsi_free_q_vectors_locked(vsi);
+
+	if (dev)
+		netdev_unlock(dev);
+}
+
 /**
  * ice_cfg_tstamp - Configure Tx time stamp queue
  * @tx_ring: Tx ring to be configured with timestamping
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 041278caf8e3..f647e583d198 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2308,10 +2308,14 @@ static int ice_vsi_cfg_tc_lan(struct ice_pf *pf, struct ice_vsi *vsi)
 }
 
 /**
- * ice_vsi_cfg_def - configure default VSI based on the type
+ * ice_vsi_cfg_def_locked - configure default VSI based on the type
  * @vsi: pointer to VSI
+ *
+ * Should be called only with the netdev lock taken.
+ *
+ * Return: 0 on success, -errno on failure.
  */
-static int ice_vsi_cfg_def(struct ice_vsi *vsi)
+static int ice_vsi_cfg_def_locked(struct ice_vsi *vsi)
 {
 	struct device *dev = ice_pf_to_dev(vsi->back);
 	struct ice_pf *pf = vsi->back;
@@ -2354,7 +2358,7 @@ static int ice_vsi_cfg_def(struct ice_vsi *vsi)
 	case ICE_VSI_CTRL:
 	case ICE_VSI_SF:
 	case ICE_VSI_PF:
-		ret = ice_vsi_alloc_q_vectors(vsi);
+		ret = ice_vsi_alloc_q_vectors_locked(vsi);
 		if (ret)
 			goto unroll_vsi_init;
 
@@ -2404,7 +2408,7 @@ static int ice_vsi_cfg_def(struct ice_vsi *vsi)
 		 * creates a VSI and corresponding structures for bookkeeping
 		 * purpose
 		 */
-		ret = ice_vsi_alloc_q_vectors(vsi);
+		ret = ice_vsi_alloc_q_vectors_locked(vsi);
 		if (ret)
 			goto unroll_vsi_init;
 
@@ -2460,6 +2464,28 @@ static int ice_vsi_cfg_def(struct ice_vsi *vsi)
 	return ret;
 }
 
+/**
+ * ice_vsi_cfg_def - configure default VSI based on the type
+ * @vsi: pointer to VSI
+ *
+ * Return: 0 on success, -errno on failure.
+ */
+static int ice_vsi_cfg_def(struct ice_vsi *vsi)
+{
+	struct net_device *dev = vsi->netdev;
+	int ret;
+
+	if (dev)
+		netdev_lock(dev);
+
+	ret = ice_vsi_cfg_def_locked(vsi);
+
+	if (dev)
+		netdev_unlock(dev);
+
+	return ret;
+}
+
 /**
  * ice_vsi_cfg - configure a previously allocated VSI
  * @vsi: pointer to VSI
@@ -2494,10 +2520,12 @@ int ice_vsi_cfg(struct ice_vsi *vsi)
 }
 
 /**
- * ice_vsi_decfg - remove all VSI configuration
+ * ice_vsi_decfg_locked - remove all VSI configuration
  * @vsi: pointer to VSI
+ *
+ * Should be called only under the netdev lock.
  */
-void ice_vsi_decfg(struct ice_vsi *vsi)
+static void ice_vsi_decfg_locked(struct ice_vsi *vsi)
 {
 	struct ice_pf *pf = vsi->back;
 	int err;
@@ -2515,7 +2543,7 @@ void ice_vsi_decfg(struct ice_vsi *vsi)
 		ice_destroy_xdp_rings(vsi, ICE_XDP_CFG_PART);
 
 	ice_vsi_clear_rings(vsi);
-	ice_vsi_free_q_vectors(vsi);
+	ice_vsi_free_q_vectors_locked(vsi);
 	ice_vsi_put_qs(vsi);
 	ice_vsi_free_arrays(vsi);
 
@@ -2530,6 +2558,23 @@ void ice_vsi_decfg(struct ice_vsi *vsi)
 		vsi->agg_node->num_vsis--;
 }
 
+/**
+ * ice_vsi_decfg - remove all VSI configuration
+ * @vsi: pointer to VSI
+ */
+void ice_vsi_decfg(struct ice_vsi *vsi)
+{
+	struct net_device *dev = vsi->netdev;
+
+	if (dev)
+		netdev_lock(dev);
+
+	ice_vsi_decfg_locked(vsi);
+
+	if (dev)
+		netdev_unlock(dev);
+}
+
 /**
  * ice_vsi_setup - Set up a VSI by a given type
  * @pf: board private structure
@@ -3107,16 +3152,17 @@ ice_vsi_realloc_stat_arrays(struct ice_vsi *vsi)
 }
 
 /**
- * ice_vsi_rebuild - Rebuild VSI after reset
+ * ice_vsi_rebuild_locked - Rebuild VSI after reset
  * @vsi: VSI to be rebuild
  * @vsi_flags: flags used for VSI rebuild flow
  *
  * Set vsi_flags to ICE_VSI_FLAG_INIT to initialize a new VSI, or
  * ICE_VSI_FLAG_NO_INIT to rebuild an existing VSI in hardware.
+ * Should be called only under the netdev lock.
  *
  * Returns 0 on success and negative value on failure
  */
-int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags)
+int ice_vsi_rebuild_locked(struct ice_vsi *vsi, u32 vsi_flags)
 {
 	struct ice_coalesce_stored *coalesce;
 	int prev_num_q_vectors;
@@ -3137,8 +3183,8 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags)
 	if (ret)
 		goto unlock;
 
-	ice_vsi_decfg(vsi);
-	ret = ice_vsi_cfg_def(vsi);
+	ice_vsi_decfg_locked(vsi);
+	ret = ice_vsi_cfg_def_locked(vsi);
 	if (ret)
 		goto unlock;
 
@@ -3175,6 +3221,32 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags)
 	return ret;
 }
 
+/**
+ * ice_vsi_rebuild - Rebuild VSI after reset
+ * @vsi: VSI to be rebuild
+ * @vsi_flags: flags used for VSI rebuild flow
+ *
+ * Set vsi_flags to ICE_VSI_FLAG_INIT to initialize a new VSI, or
+ * ICE_VSI_FLAG_NO_INIT to rebuild an existing VSI in hardware.
+ *
+ * Return: 0 on success, -errno on failure.
+ */
+int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags)
+{
+	struct net_device *dev = vsi->netdev;
+	int ret;
+
+	if (dev)
+		netdev_lock(dev);
+
+	ret = ice_vsi_rebuild_locked(vsi, vsi_flags);
+
+	if (dev)
+		netdev_unlock(dev);
+
+	return ret;
+}
+
 /**
  * ice_is_reset_in_progress - check for a reset in progress
  * @state: PF state field
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 7a1c6859d708..b9d726797919 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -4130,6 +4130,7 @@ bool ice_is_wol_supported(struct ice_hw *hw)
  * @locked: is adev device_lock held
  *
  * Only change the number of queues if new_tx, or new_rx is non-0.
+ * Note that it should be called only with the netdev lock taken.
  *
  * Returns 0 on success.
  */
@@ -4155,7 +4156,7 @@ int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx, bool locked)
 
 	/* set for the next time the netdev is started */
 	if (!netif_running(vsi->netdev)) {
-		err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT);
+		err = ice_vsi_rebuild_locked(vsi, ICE_VSI_FLAG_NO_INIT);
 		if (err)
 			goto rebuild_err;
 		dev_dbg(ice_pf_to_dev(pf), "Link is down, queue count change happens when link is brought up\n");
@@ -4163,7 +4164,7 @@ int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx, bool locked)
 	}
 
 	ice_vsi_close(vsi);
-	err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT);
+	err = ice_vsi_rebuild_locked(vsi, ICE_VSI_FLAG_NO_INIT);
 	if (err)
 		goto rebuild_err;
 
-- 
2.52.0

RE: [PATCH iwl-next] ice: fix system hang on `ethtool -L`

[Loktionov, Aleksandr]

> -----Original Message-----
> From: Lobakin, Aleksander <aleksander.lobakin@intel.com>
> Sent: Tuesday, January 20, 2026 6:34 PM
> To: Nguyen, Anthony L <anthony.l.nguyen@intel.com>; intel-wired-
> lan@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Kitszel,
> Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn
> <andrew+netdev@lunn.ch>; David S. Miller <davem@davemloft.net>; Eric
> Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo
> Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller,
> Jacob E <jacob.e.keller@intel.com>; Loktionov, Aleksandr
> <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming
> <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org;
> linux-kernel@vger.kernel.org
> Subject: [PATCH iwl-next] ice: fix system hang on `ethtool -L`
> 
> ice_set_channels() calls ice_vsi_rebuild() under the netdev lock
> taken, but ice_vsi_rebuild() calls netif_napi_{add,del}() which take
> the same lock.
> Add ice_vsi_rebuild_locked() which uses the _locked counterparts of
> these functions and use it in ice_set_channels().
> 
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
> Hey Tony, please amend to the patch I replied to.
> ---
>  drivers/net/ethernet/intel/ice/ice_base.h |  2 +
> drivers/net/ethernet/intel/ice/ice_lib.h  |  1 +
> drivers/net/ethernet/intel/ice/ice_base.c | 63 ++++++++++++---
> drivers/net/ethernet/intel/ice/ice_lib.c  | 94 ++++++++++++++++++++---
> drivers/net/ethernet/intel/ice/ice_main.c |  5 +-
>  5 files changed, 143 insertions(+), 22 deletions(-)
> 
> diff --git a/drivers/net/ethernet/intel/ice/ice_base.h
> b/drivers/net/ethernet/intel/ice/ice_base.h
> index d28294247599..99b2c7232829 100644
> --- a/drivers/net/ethernet/intel/ice/ice_base.h
> +++ b/drivers/net/ethernet/intel/ice/ice_base.h
> @@ -12,8 +12,10 @@ int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg);
> int  ice_vsi_ctrl_one_rx_ring(struct ice_vsi *vsi, bool ena, u16
> rxq_idx, bool wait);  int ice_vsi_wait_one_rx_ring(struct ice_vsi
> *vsi, bool ena, u16 rxq_idx);
> +int ice_vsi_alloc_q_vectors_locked(struct ice_vsi *vsi);
>  int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi);  void
> ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi);
> +void ice_vsi_free_q_vectors_locked(struct ice_vsi *vsi);
>  void ice_vsi_free_q_vectors(struct ice_vsi *vsi);  int
> ice_vsi_cfg_single_txq(struct ice_vsi *vsi, struct ice_tx_ring
> **tx_rings,
>  			   u16 q_idx);
> diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h
> b/drivers/net/ethernet/intel/ice/ice_lib.h
> index 347b63e497e7..e55b72db72c4 100644
> --- a/drivers/net/ethernet/intel/ice/ice_lib.h
> +++ b/drivers/net/ethernet/intel/ice/ice_lib.h
> @@ -68,6 +68,7 @@ int ice_ena_vsi(struct ice_vsi *vsi, bool locked);
> void ice_vsi_decfg(struct ice_vsi *vsi);  void ice_dis_vsi(struct
> ice_vsi *vsi, bool locked);
> 
> +int ice_vsi_rebuild_locked(struct ice_vsi *vsi, u32 vsi_flags);
>  int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags);  int
> ice_vsi_cfg(struct ice_vsi *vsi);  struct ice_vsi
> *ice_vsi_alloc(struct ice_pf *pf); diff --git
> a/drivers/net/ethernet/intel/ice/ice_base.c
> b/drivers/net/ethernet/intel/ice/ice_base.c
> index 7097324c38f3..65e19815bec5 100644
> --- a/drivers/net/ethernet/intel/ice/ice_base.c
> +++ b/drivers/net/ethernet/intel/ice/ice_base.c
> @@ -153,8 +153,8 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi
> *vsi, u16 v_idx)
>  	 * handler here (i.e. resume, reset/rebuild, etc.)
>  	 */
>  	if (vsi->netdev)
> -		netif_napi_add_config(vsi->netdev, &q_vector->napi,
> -				      ice_napi_poll, v_idx);
> +		netif_napi_add_config_locked(vsi->netdev, &q_vector-
> >napi,
> +					     ice_napi_poll, v_idx);
If you converted ice_vsi_alloc_q_vector() into _locked, should it be lockdep_assert_held(&vsi->netdev->lock); then?

Everything else looks fine.
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>


...
 
> --
> 2.52.0

Re: [PATCH iwl-next] ice: fix system hang on `ethtool -L`

[Alexander Lobakin]

From: Loktionov, Aleksandr <aleksandr.loktionov@intel.com>
Date: Wed, 21 Jan 2026 08:18:47 +0100

> 
> 
>> -----Original Message-----
>> From: Lobakin, Aleksander <aleksander.lobakin@intel.com>
>> Sent: Tuesday, January 20, 2026 6:34 PM
>> To: Nguyen, Anthony L <anthony.l.nguyen@intel.com>; intel-wired-
>> lan@lists.osuosl.org
>> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Kitszel,
>> Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn
>> <andrew+netdev@lunn.ch>; David S. Miller <davem@davemloft.net>; Eric
>> Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo
>> Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller,
>> Jacob E <jacob.e.keller@intel.com>; Loktionov, Aleksandr
>> <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming
>> <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org;
>> linux-kernel@vger.kernel.org
>> Subject: [PATCH iwl-next] ice: fix system hang on `ethtool -L`
>>
>> ice_set_channels() calls ice_vsi_rebuild() under the netdev lock
>> taken, but ice_vsi_rebuild() calls netif_napi_{add,del}() which take
>> the same lock.
>> Add ice_vsi_rebuild_locked() which uses the _locked counterparts of
>> these functions and use it in ice_set_channels().
>>
>> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
>> ---
>> Hey Tony, please amend to the patch I replied to.
>> ---
>>  drivers/net/ethernet/intel/ice/ice_base.h |  2 +
>> drivers/net/ethernet/intel/ice/ice_lib.h  |  1 +
>> drivers/net/ethernet/intel/ice/ice_base.c | 63 ++++++++++++---
>> drivers/net/ethernet/intel/ice/ice_lib.c  | 94 ++++++++++++++++++++---
>> drivers/net/ethernet/intel/ice/ice_main.c |  5 +-
>>  5 files changed, 143 insertions(+), 22 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/intel/ice/ice_base.h
>> b/drivers/net/ethernet/intel/ice/ice_base.h
>> index d28294247599..99b2c7232829 100644
>> --- a/drivers/net/ethernet/intel/ice/ice_base.h
>> +++ b/drivers/net/ethernet/intel/ice/ice_base.h
>> @@ -12,8 +12,10 @@ int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg);
>> int  ice_vsi_ctrl_one_rx_ring(struct ice_vsi *vsi, bool ena, u16
>> rxq_idx, bool wait);  int ice_vsi_wait_one_rx_ring(struct ice_vsi
>> *vsi, bool ena, u16 rxq_idx);
>> +int ice_vsi_alloc_q_vectors_locked(struct ice_vsi *vsi);
>>  int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi);  void
>> ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi);
>> +void ice_vsi_free_q_vectors_locked(struct ice_vsi *vsi);
>>  void ice_vsi_free_q_vectors(struct ice_vsi *vsi);  int
>> ice_vsi_cfg_single_txq(struct ice_vsi *vsi, struct ice_tx_ring
>> **tx_rings,
>>  			   u16 q_idx);
>> diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h
>> b/drivers/net/ethernet/intel/ice/ice_lib.h
>> index 347b63e497e7..e55b72db72c4 100644
>> --- a/drivers/net/ethernet/intel/ice/ice_lib.h
>> +++ b/drivers/net/ethernet/intel/ice/ice_lib.h
>> @@ -68,6 +68,7 @@ int ice_ena_vsi(struct ice_vsi *vsi, bool locked);
>> void ice_vsi_decfg(struct ice_vsi *vsi);  void ice_dis_vsi(struct
>> ice_vsi *vsi, bool locked);
>>
>> +int ice_vsi_rebuild_locked(struct ice_vsi *vsi, u32 vsi_flags);
>>  int ice_vsi_rebuild(struct ice_vsi *vsi, u32 vsi_flags);  int
>> ice_vsi_cfg(struct ice_vsi *vsi);  struct ice_vsi
>> *ice_vsi_alloc(struct ice_pf *pf); diff --git
>> a/drivers/net/ethernet/intel/ice/ice_base.c
>> b/drivers/net/ethernet/intel/ice/ice_base.c
>> index 7097324c38f3..65e19815bec5 100644
>> --- a/drivers/net/ethernet/intel/ice/ice_base.c
>> +++ b/drivers/net/ethernet/intel/ice/ice_base.c
>> @@ -153,8 +153,8 @@ static int ice_vsi_alloc_q_vector(struct ice_vsi
>> *vsi, u16 v_idx)
>>  	 * handler here (i.e. resume, reset/rebuild, etc.)
>>  	 */
>>  	if (vsi->netdev)
>> -		netif_napi_add_config(vsi->netdev, &q_vector->napi,
>> -				      ice_napi_poll, v_idx);
>> +		netif_napi_add_config_locked(vsi->netdev, &q_vector-
>>> napi,
>> +					     ice_napi_poll, v_idx);
> If you converted ice_vsi_alloc_q_vector() into _locked, should it be lockdep_assert_held(&vsi->netdev->lock); then?

IIRC the core kernel functions check for this and warn already.

> 
> Everything else looks fine.
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>

Thanks,
Olek

RE: [Intel-wired-lan] [PATCH iwl-next v2 3/5] ice: migrate to netdev ops lock

[Nowlin, Alexander]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Alexander Lobakin
> Sent: Thursday, December 4, 2025 7:52 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Kitszel, Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn <andrew+netdev@lunn.ch>; 
> David S. Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller, Jacob E 
> <jacob.e.keller@intel.com>; Loktionov, Aleksandr <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org; linux-
> kernel@vger.kernel.org
> Subject: [Intel-wired-lan] [PATCH iwl-next v2 3/5] ice: migrate to netdev ops lock
> 
> Queue management ops unconditionally enable netdev locking. The same lock is taken by default by several NAPI configuration functions, such as napi_enable() and netif_napi_set_irq().
> Request ops locking in advance and make sure we use the _locked counterparts of those functions to avoid deadlocks, taking the lock manually where needed (suspend/resume, queue rebuild and resets).
> 
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_lib.h    |  6 ++-
>  drivers/net/ethernet/intel/ice/ice_lib.c    | 56 +++++++++++++++++----
>  drivers/net/ethernet/intel/ice/ice_main.c   | 49 ++++++++++--------
>  drivers/net/ethernet/intel/ice/ice_sf_eth.c |  1 +
>  drivers/net/ethernet/intel/ice/ice_xsk.c    |  4 +-
>  5 files changed, 82 insertions(+), 34 deletions(-)

Tested-by: Alexander Nowlin <alexander.nowlin@intel.com>

[PATCH iwl-next v2 4/5] ice: implement Rx queue management ops

[Alexander Lobakin]

Now ice is ready to get queue_mgmt_ops support. It already has API
to disable/reconfig/enable one particular queue (for XSk). Reuse as
much of its code as possible to implement Rx queue management
callbacks and vice versa -- ice_queue_mem_{alloc,free}() can be
reused during ifup/ifdown to elide code duplication.
With this, ice passes the io_uring zcrx selftests, meaning the Rx
part of netmem/MP support is done.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_lib.h    |   5 +
 drivers/net/ethernet/intel/ice/ice_txrx.h   |   2 +
 drivers/net/ethernet/intel/ice/ice_base.c   | 192 ++++++++++++++------
 drivers/net/ethernet/intel/ice/ice_main.c   |   2 +-
 drivers/net/ethernet/intel/ice/ice_sf_eth.c |   2 +-
 drivers/net/ethernet/intel/ice/ice_txrx.c   |  26 ++-
 6 files changed, 163 insertions(+), 66 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h
index 347b63e497e7..2d3168458891 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_lib.h
@@ -4,6 +4,8 @@
 #ifndef _ICE_LIB_H_
 #define _ICE_LIB_H_
 
+#include <net/netdev_queues.h>
+
 #include "ice.h"
 #include "ice_vlan.h"
 
@@ -132,4 +134,7 @@ void ice_clear_feature_support(struct ice_pf *pf, enum ice_feature f);
 void ice_init_feature_support(struct ice_pf *pf);
 bool ice_vsi_is_rx_queue_active(struct ice_vsi *vsi);
 void ice_vsi_update_l2tsel(struct ice_vsi *vsi, enum ice_l2tsel l2tsel);
+
+extern const struct netdev_queue_mgmt_ops ice_queue_mgmt_ops;
+
 #endif /* !_ICE_LIB_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index c51b1e60f717..e3c682723107 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -440,6 +440,8 @@ u16
 ice_select_queue(struct net_device *dev, struct sk_buff *skb,
 		 struct net_device *sb_dev);
 void ice_clean_tx_ring(struct ice_tx_ring *tx_ring);
+void ice_queue_mem_free(struct net_device *dev, void *per_queue_mem);
+void ice_zero_rx_ring(struct ice_rx_ring *rx_ring);
 void ice_clean_rx_ring(struct ice_rx_ring *rx_ring);
 int ice_setup_tx_ring(struct ice_tx_ring *tx_ring);
 int ice_setup_rx_ring(struct ice_rx_ring *rx_ring);
diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index 7097324c38f3..4479bb5e66ac 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -651,6 +651,42 @@ static int ice_rxq_pp_create(struct ice_rx_ring *rq)
 	return err;
 }
 
+static int ice_queue_mem_alloc(struct net_device *dev, void *per_queue_mem,
+			       int idx)
+{
+	const struct ice_netdev_priv *priv = netdev_priv(dev);
+	const struct ice_rx_ring *real = priv->vsi->rx_rings[idx];
+	struct ice_rx_ring *new = per_queue_mem;
+	int ret;
+
+	new->count = real->count;
+	new->netdev = real->netdev;
+	new->q_index = real->q_index;
+	new->q_vector = real->q_vector;
+	new->vsi = real->vsi;
+
+	ret = ice_rxq_pp_create(new);
+	if (ret)
+		return ret;
+
+	if (!netif_running(dev))
+		return 0;
+
+	ret = __xdp_rxq_info_reg(&new->xdp_rxq, new->netdev, new->q_index,
+				 new->q_vector->napi.napi_id, new->rx_buf_len);
+	if (ret)
+		goto err_destroy_fq;
+
+	xdp_rxq_info_attach_page_pool(&new->xdp_rxq, new->pp);
+
+	return 0;
+
+err_destroy_fq:
+	ice_rxq_pp_destroy(new);
+
+	return ret;
+}
+
 /**
  * ice_vsi_cfg_rxq - Configure an Rx queue
  * @ring: the ring being configured
@@ -665,23 +701,12 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring)
 	int err;
 
 	if (ring->vsi->type == ICE_VSI_PF || ring->vsi->type == ICE_VSI_SF) {
-		if (!xdp_rxq_info_is_reg(&ring->xdp_rxq)) {
-			err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
-						 ring->q_index,
-						 ring->q_vector->napi.napi_id,
-						 ring->rx_buf_len);
-			if (err)
-				return err;
-		}
-
 		ice_rx_xsk_pool(ring);
 		err = ice_realloc_rx_xdp_bufs(ring, ring->xsk_pool);
 		if (err)
 			return err;
 
 		if (ring->xsk_pool) {
-			xdp_rxq_info_unreg(&ring->xdp_rxq);
-
 			rx_buf_len =
 				xsk_pool_get_rx_frame_size(ring->xsk_pool);
 			err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
@@ -700,20 +725,10 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring)
 			dev_info(dev, "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n",
 				 ring->q_index);
 		} else {
-			err = ice_rxq_pp_create(ring);
+			err = ice_queue_mem_alloc(ring->netdev, ring,
+						  ring->q_index);
 			if (err)
 				return err;
-
-			if (!xdp_rxq_info_is_reg(&ring->xdp_rxq)) {
-				err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev,
-							 ring->q_index,
-							 ring->q_vector->napi.napi_id,
-							 ring->rx_buf_len);
-				if (err)
-					goto err_destroy_fq;
-			}
-			xdp_rxq_info_attach_page_pool(&ring->xdp_rxq,
-						      ring->pp);
 		}
 	}
 
@@ -722,7 +737,7 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring)
 	if (err) {
 		dev_err(dev, "ice_setup_rx_ctx failed for RxQ %d, err %d\n",
 			ring->q_index, err);
-		goto err_destroy_fq;
+		goto err_clean_rq;
 	}
 
 	if (ring->xsk_pool) {
@@ -753,12 +768,12 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring)
 		err = ice_alloc_rx_bufs(ring, num_bufs);
 
 	if (err)
-		goto err_destroy_fq;
+		goto err_clean_rq;
 
 	return 0;
 
-err_destroy_fq:
-	ice_rxq_pp_destroy(ring);
+err_clean_rq:
+	ice_clean_rx_ring(ring);
 
 	return err;
 }
@@ -1425,27 +1440,7 @@ static void ice_qp_reset_stats(struct ice_vsi *vsi, u16 q_idx)
 		       sizeof(vsi->xdp_rings[q_idx]->ring_stats->stats));
 }
 
-/**
- * ice_qp_clean_rings - Cleans all the rings of a given index
- * @vsi: VSI that contains rings of interest
- * @q_idx: ring index in array
- */
-static void ice_qp_clean_rings(struct ice_vsi *vsi, u16 q_idx)
-{
-	ice_clean_tx_ring(vsi->tx_rings[q_idx]);
-	if (vsi->xdp_rings)
-		ice_clean_tx_ring(vsi->xdp_rings[q_idx]);
-	ice_clean_rx_ring(vsi->rx_rings[q_idx]);
-}
-
-/**
- * ice_qp_dis - Disables a queue pair
- * @vsi: VSI of interest
- * @q_idx: ring index in array
- *
- * Returns 0 on success, negative on failure.
- */
-int ice_qp_dis(struct ice_vsi *vsi, u16 q_idx)
+static int __ice_qp_dis(struct ice_vsi *vsi, u16 q_idx)
 {
 	struct ice_txq_meta txq_meta = { };
 	struct ice_q_vector *q_vector;
@@ -1484,23 +1479,35 @@ int ice_qp_dis(struct ice_vsi *vsi, u16 q_idx)
 	}
 
 	ice_vsi_ctrl_one_rx_ring(vsi, false, q_idx, false);
-	ice_qp_clean_rings(vsi, q_idx);
 	ice_qp_reset_stats(vsi, q_idx);
 
+	ice_clean_tx_ring(vsi->tx_rings[q_idx]);
+	if (vsi->xdp_rings)
+		ice_clean_tx_ring(vsi->xdp_rings[q_idx]);
+
 	return fail;
 }
 
 /**
- * ice_qp_ena - Enables a queue pair
+ * ice_qp_dis - Disables a queue pair
  * @vsi: VSI of interest
  * @q_idx: ring index in array
  *
  * Returns 0 on success, negative on failure.
  */
-int ice_qp_ena(struct ice_vsi *vsi, u16 q_idx)
+int ice_qp_dis(struct ice_vsi *vsi, u16 q_idx)
+{
+	int ret;
+
+	ret = __ice_qp_dis(vsi, q_idx);
+	ice_clean_rx_ring(vsi->rx_rings[q_idx]);
+
+	return ret;
+}
+
+static int __ice_qp_ena(struct ice_vsi *vsi, u16 q_idx, int fail)
 {
 	struct ice_q_vector *q_vector;
-	int fail = 0;
 	bool link_up;
 	int err;
 
@@ -1518,10 +1525,6 @@ int ice_qp_ena(struct ice_vsi *vsi, u16 q_idx)
 		ice_tx_xsk_pool(vsi, q_idx);
 	}
 
-	err = ice_vsi_cfg_single_rxq(vsi, q_idx);
-	if (!fail)
-		fail = err;
-
 	q_vector = vsi->rx_rings[q_idx]->q_vector;
 	ice_qvec_cfg_msix(vsi, q_vector, q_idx);
 
@@ -1542,3 +1545,80 @@ int ice_qp_ena(struct ice_vsi *vsi, u16 q_idx)
 
 	return fail;
 }
+
+/**
+ * ice_qp_ena - Enables a queue pair
+ * @vsi: VSI of interest
+ * @q_idx: ring index in array
+ *
+ * Returns 0 on success, negative on failure.
+ */
+int ice_qp_ena(struct ice_vsi *vsi, u16 q_idx)
+{
+	return __ice_qp_ena(vsi, q_idx, ice_vsi_cfg_single_rxq(vsi, q_idx));
+}
+
+static int ice_queue_start(struct net_device *dev, void *per_queue_mem,
+			   int idx)
+{
+	const struct ice_netdev_priv *priv = netdev_priv(dev);
+	struct ice_rx_ring *real = priv->vsi->rx_rings[idx];
+	struct ice_rx_ring *new = per_queue_mem;
+	struct napi_struct *napi;
+	int ret;
+
+	real->pp = new->pp;
+	real->rx_fqes = new->rx_fqes;
+	real->hdr_fqes = new->hdr_fqes;
+	real->hdr_pp = new->hdr_pp;
+
+	real->hdr_truesize = new->hdr_truesize;
+	real->truesize = new->truesize;
+	real->rx_hdr_len = new->rx_hdr_len;
+	real->rx_buf_len = new->rx_buf_len;
+
+	memcpy(&real->xdp_rxq, &new->xdp_rxq, sizeof(new->xdp_rxq));
+
+	ret = ice_setup_rx_ctx(real);
+	if (ret)
+		return ret;
+
+	napi = &real->q_vector->napi;
+
+	page_pool_enable_direct_recycling(real->pp, napi);
+	if (real->hdr_pp)
+		page_pool_enable_direct_recycling(real->hdr_pp, napi);
+
+	ret = ice_alloc_rx_bufs(real, ICE_DESC_UNUSED(real));
+
+	return __ice_qp_ena(priv->vsi, idx, ret);
+}
+
+static int ice_queue_stop(struct net_device *dev, void *per_queue_mem,
+			  int idx)
+{
+	const struct ice_netdev_priv *priv = netdev_priv(dev);
+	struct ice_rx_ring *real = priv->vsi->rx_rings[idx];
+	int ret;
+
+	ret = __ice_qp_dis(priv->vsi, idx);
+	if (ret)
+		return ret;
+
+	page_pool_disable_direct_recycling(real->pp);
+	if (real->hdr_pp)
+		page_pool_disable_direct_recycling(real->hdr_pp);
+
+	ice_zero_rx_ring(real);
+	memcpy(per_queue_mem, real, sizeof(*real));
+
+	return 0;
+}
+
+const struct netdev_queue_mgmt_ops ice_queue_mgmt_ops = {
+	.ndo_queue_mem_alloc	= ice_queue_mem_alloc,
+	.ndo_queue_mem_free	= ice_queue_mem_free,
+	.ndo_queue_mem_size	= sizeof(struct ice_rx_ring),
+	.ndo_queue_start	= ice_queue_start,
+	.ndo_queue_stop		= ice_queue_stop,
+};
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 881b41d37821..25e9091ca309 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3523,7 +3523,7 @@ static void ice_set_ops(struct ice_vsi *vsi)
 	}
 
 	netdev->netdev_ops = &ice_netdev_ops;
-	netdev->request_ops_lock = true;
+	netdev->queue_mgmt_ops = &ice_queue_mgmt_ops;
 	netdev->udp_tunnel_nic_info = &pf->hw.udp_tunnel_nic;
 	netdev->xdp_metadata_ops = &ice_xdp_md_ops;
 	ice_set_ethtool_ops(netdev);
diff --git a/drivers/net/ethernet/intel/ice/ice_sf_eth.c b/drivers/net/ethernet/intel/ice/ice_sf_eth.c
index 2c3db1b03055..41e1606a8222 100644
--- a/drivers/net/ethernet/intel/ice/ice_sf_eth.c
+++ b/drivers/net/ethernet/intel/ice/ice_sf_eth.c
@@ -58,7 +58,7 @@ static int ice_sf_cfg_netdev(struct ice_dynamic_port *dyn_port,
 	eth_hw_addr_set(netdev, dyn_port->hw_addr);
 	ether_addr_copy(netdev->perm_addr, dyn_port->hw_addr);
 	netdev->netdev_ops = &ice_sf_netdev_ops;
-	netdev->request_ops_lock = true;
+	netdev->queue_mgmt_ops = &ice_queue_mgmt_ops;
 	SET_NETDEV_DEVLINK_PORT(netdev, devlink_port);
 
 	err = register_netdev(netdev);
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index 47d44796635a..b6f56cb81f93 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -530,17 +530,13 @@ void ice_rxq_pp_destroy(struct ice_rx_ring *rq)
 	rq->hdr_pp = NULL;
 }
 
-/**
- * ice_clean_rx_ring - Free Rx buffers
- * @rx_ring: ring to be cleaned
- */
-void ice_clean_rx_ring(struct ice_rx_ring *rx_ring)
+void ice_queue_mem_free(struct net_device *dev, void *per_queue_mem)
 {
-	u32 size;
+	struct ice_rx_ring *rx_ring = per_queue_mem;
 
 	if (rx_ring->xsk_pool) {
 		ice_xsk_clean_rx_ring(rx_ring);
-		goto rx_skip_free;
+		return;
 	}
 
 	/* ring already cleared, nothing to do */
@@ -567,8 +563,12 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring)
 	}
 
 	ice_rxq_pp_destroy(rx_ring);
+}
+
+void ice_zero_rx_ring(struct ice_rx_ring *rx_ring)
+{
+	size_t size;
 
-rx_skip_free:
 	/* Zero out the descriptor ring */
 	size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc),
 		     PAGE_SIZE);
@@ -579,6 +579,16 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring)
 	rx_ring->next_to_use = 0;
 }
 
+/**
+ * ice_clean_rx_ring - Free Rx buffers
+ * @rx_ring: ring to be cleaned
+ */
+void ice_clean_rx_ring(struct ice_rx_ring *rx_ring)
+{
+	ice_queue_mem_free(rx_ring->netdev, rx_ring);
+	ice_zero_rx_ring(rx_ring);
+}
+
 /**
  * ice_free_rx_ring - Free Rx resources
  * @rx_ring: ring to clean the resources from
-- 
2.52.0

RE: [Intel-wired-lan] [PATCH iwl-next v2 4/5] ice: implement Rx queue management ops

[Nowlin, Alexander]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Alexander Lobakin
> Sent: Thursday, December 4, 2025 7:52 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Kitszel, Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn <andrew+netdev@lunn.ch>; 
> David S. Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller, Jacob E 
> <jacob.e.keller@intel.com>; Loktionov, Aleksandr <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org; linux-
> kernel@vger.kernel.org
> Subject: [Intel-wired-lan] [PATCH iwl-next v2 4/5] ice: implement Rx queue management ops
> 
> Now ice is ready to get queue_mgmt_ops support. It already has API to disable/reconfig/enable one particular queue (for XSk). Reuse as much of its code as possible to implement Rx queue management
> callbacks and vice versa -- ice_queue_mem_{alloc,free}() can be reused during ifup/ifdown to elide code duplication.
> With this, ice passes the io_uring zcrx selftests, meaning the Rx part of netmem/MP support is done.
> 
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_lib.h    |   5 +
>  drivers/net/ethernet/intel/ice/ice_txrx.h   |   2 +
>  drivers/net/ethernet/intel/ice/ice_base.c   | 192 ++++++++++++++------
>  drivers/net/ethernet/intel/ice/ice_main.c   |   2 +-
>  drivers/net/ethernet/intel/ice/ice_sf_eth.c |   2 +-
>  drivers/net/ethernet/intel/ice/ice_txrx.c   |  26 ++-
>  6 files changed, 163 insertions(+), 66 deletions(-)

Tested-by: Alexander Nowlin <alexander.nowlin@intel.com>

[PATCH iwl-next v2 5/5] ice: add support for transmitting unreadable frags

[Alexander Lobakin]

Advertise netmem Tx support in ice. The only change needed is to set
ICE_TX_BUF_FRAG conditionally, only when skb_frag_is_net_iov() is
false. Otherwise, the Tx buffer type will be ICE_TX_BUF_EMPTY and
the driver will skip the DMA unmapping operation.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c   |  1 +
 drivers/net/ethernet/intel/ice/ice_sf_eth.c |  1 +
 drivers/net/ethernet/intel/ice/ice_txrx.c   | 17 +++++++++++++----
 3 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 25e9091ca309..66601b1b7fec 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3524,6 +3524,7 @@ static void ice_set_ops(struct ice_vsi *vsi)
 
 	netdev->netdev_ops = &ice_netdev_ops;
 	netdev->queue_mgmt_ops = &ice_queue_mgmt_ops;
+	netdev->netmem_tx = true;
 	netdev->udp_tunnel_nic_info = &pf->hw.udp_tunnel_nic;
 	netdev->xdp_metadata_ops = &ice_xdp_md_ops;
 	ice_set_ethtool_ops(netdev);
diff --git a/drivers/net/ethernet/intel/ice/ice_sf_eth.c b/drivers/net/ethernet/intel/ice/ice_sf_eth.c
index 41e1606a8222..51ad13c9d7f9 100644
--- a/drivers/net/ethernet/intel/ice/ice_sf_eth.c
+++ b/drivers/net/ethernet/intel/ice/ice_sf_eth.c
@@ -59,6 +59,7 @@ static int ice_sf_cfg_netdev(struct ice_dynamic_port *dyn_port,
 	ether_addr_copy(netdev->perm_addr, dyn_port->hw_addr);
 	netdev->netdev_ops = &ice_sf_netdev_ops;
 	netdev->queue_mgmt_ops = &ice_queue_mgmt_ops;
+	netdev->netmem_tx = true;
 	SET_NETDEV_DEVLINK_PORT(netdev, devlink_port);
 
 	err = register_netdev(netdev);
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index b6f56cb81f93..e8e1acbd5a7d 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -113,11 +113,17 @@ ice_prgm_fdir_fltr(struct ice_vsi *vsi, struct ice_fltr_desc *fdir_desc,
 static void
 ice_unmap_and_free_tx_buf(struct ice_tx_ring *ring, struct ice_tx_buf *tx_buf)
 {
-	if (tx_buf->type != ICE_TX_BUF_XDP_TX && dma_unmap_len(tx_buf, len))
+	switch (tx_buf->type) {
+	case ICE_TX_BUF_DUMMY:
+	case ICE_TX_BUF_FRAG:
+	case ICE_TX_BUF_SKB:
+	case ICE_TX_BUF_XDP_XMIT:
 		dma_unmap_page(ring->dev,
 			       dma_unmap_addr(tx_buf, dma),
 			       dma_unmap_len(tx_buf, len),
 			       DMA_TO_DEVICE);
+		break;
+	}
 
 	switch (tx_buf->type) {
 	case ICE_TX_BUF_DUMMY:
@@ -337,12 +343,14 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 			}
 
 			/* unmap any remaining paged data */
-			if (dma_unmap_len(tx_buf, len)) {
+			if (tx_buf->type != ICE_TX_BUF_EMPTY) {
 				dma_unmap_page(tx_ring->dev,
 					       dma_unmap_addr(tx_buf, dma),
 					       dma_unmap_len(tx_buf, len),
 					       DMA_TO_DEVICE);
+
 				dma_unmap_len_set(tx_buf, len, 0);
+				tx_buf->type = ICE_TX_BUF_EMPTY;
 			}
 		}
 		ice_trace(clean_tx_irq_unmap_eop, tx_ring, tx_desc, tx_buf);
@@ -1493,7 +1501,8 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first,
 				       DMA_TO_DEVICE);
 
 		tx_buf = &tx_ring->tx_buf[i];
-		tx_buf->type = ICE_TX_BUF_FRAG;
+		if (!skb_frag_is_net_iov(frag))
+			tx_buf->type = ICE_TX_BUF_FRAG;
 	}
 
 	/* record SW timestamp if HW timestamp is not available */
@@ -2368,7 +2377,7 @@ void ice_clean_ctrl_tx_irq(struct ice_tx_ring *tx_ring)
 		}
 
 		/* unmap the data header */
-		if (dma_unmap_len(tx_buf, len))
+		if (tx_buf->type != ICE_TX_BUF_EMPTY)
 			dma_unmap_single(tx_ring->dev,
 					 dma_unmap_addr(tx_buf, dma),
 					 dma_unmap_len(tx_buf, len),
-- 
2.52.0

RE: [Intel-wired-lan] [PATCH iwl-next v2 5/5] ice: add support for transmitting unreadable frags

[Nowlin, Alexander]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Alexander Lobakin
> Sent: Thursday, December 4, 2025 7:52 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.lobakin@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Kitszel, Przemyslaw <przemyslaw.kitszel@intel.com>; Andrew Lunn <andrew+netdev@lunn.ch>;
> David S. Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>; Jakub Kicinski <kuba@kernel.org>; Paolo Abeni <pabeni@redhat.com>; Simon Horman <horms@kernel.org>; Keller, Jacob E
> <jacob.e.keller@intel.com>; Loktionov, Aleksandr <aleksandr.loktionov@intel.com>; NXNE CNSE OSDT ITP Upstreaming <nxne.cnse.osdt.itp.upstreaming@intel.com>; netdev@vger.kernel.org; linux-
> kernel@vger.kernel.org
> Subject: [Intel-wired-lan] [PATCH iwl-next v2 5/5] ice: add support for transmitting unreadable frags
> 
> Advertise netmem Tx support in ice. The only change needed is to set ICE_TX_BUF_FRAG conditionally, only when skb_frag_is_net_iov() is false. Otherwise, the Tx buffer type will be ICE_TX_BUF_EMPTY and the > driver will skip the DMA unmapping operation.
> 
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_main.c   |  1 +
>  drivers/net/ethernet/intel/ice/ice_sf_eth.c |  1 +
>  drivers/net/ethernet/intel/ice/ice_txrx.c   | 17 +++++++++++++----
>  3 files changed, 15 insertions(+), 4 deletions(-)

Tested-by: Alexander Nowlin <alexander.nowlin@intel.com>

Re: [PATCH iwl-next v2 0/5] ice: add support for devmem/io_uring Rx and Tx

[Alexander Lobakin]

From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 4 Dec 2025 16:51:28 +0100

> Now that ice uses libeth for managing Rx buffers and supports
> configurable header split, it's ready to get support for sending
> and receiving packets with unreadable (to the kernel) frags.
> 
> Extend libeth just a little bit to allow creating PPs with custom
> memory providers and make sure ice works correctly with the netdev
> ops locking. Then add the full set of queue_mgmt_ops and don't
> unmap unreadable frags on Tx completion.
> No perf regressions for the regular flows and no code duplication
> implied.
> 
> Credits to the fbnic developers, which's code helped me understand
> the memory providers and queue_mgmt_ops logics and served as
> a reference.
> 
> Alexander Lobakin (5):
>   libeth: pass Rx queue index to PP when creating a fill queue
>   libeth: handle creating pools with unreadable buffers
>   ice: migrate to netdev ops lock
>   ice: implement Rx queue management ops
>   ice: add support for transmitting unreadable frags
> 
>  drivers/net/ethernet/intel/ice/ice_lib.h    |  11 +-
>  drivers/net/ethernet/intel/ice/ice_txrx.h   |   2 +
>  drivers/net/ethernet/intel/idpf/idpf_txrx.h |   2 +
>  include/net/libeth/rx.h                     |   2 +
>  include/net/libeth/tx.h                     |   2 +-
>  drivers/net/ethernet/intel/iavf/iavf_txrx.c |   1 +
>  drivers/net/ethernet/intel/ice/ice_base.c   | 194 ++++++++++++++------
>  drivers/net/ethernet/intel/ice/ice_lib.c    |  56 +++++-
>  drivers/net/ethernet/intel/ice/ice_main.c   |  50 ++---
>  drivers/net/ethernet/intel/ice/ice_sf_eth.c |   2 +
>  drivers/net/ethernet/intel/ice/ice_txrx.c   |  43 +++--
>  drivers/net/ethernet/intel/ice/ice_xsk.c    |   4 +-
>  drivers/net/ethernet/intel/idpf/idpf_txrx.c |  13 ++
>  drivers/net/ethernet/intel/libeth/rx.c      |  46 +++++
>  14 files changed, 325 insertions(+), 103 deletions(-)
> 
> ---
> From v1[0]:
> * rebase on top of the latest next-queue;
> * fix a typo 'rxq_ixd' -> 'rxq_idx' (Tony).
> 
> Testing hints:
> * regular Rx and Tx for regressions;
> * <kernel root>/tools/testing/selftests/drivers/net/hw/ contains
>   scripts for testing netmem Rx and Tx, namely devmem.py and
>   iou-zcrx.py (read the documentation first).
> 
> [0] https://lore.kernel.org/intel-wired-lan/20251125173603.3834486-1-aleksander.lobakin@intel.com

Current status from the validation:

* io_uring tests pass, except for tests which try to setup RSS Rx
  contexts -- after recent changes in the kernel, seems like at least
  some of our drivers return -EOPNOTSUPP when trying to configure RSS.

Tony, did you have any related reports or planned fixes?

* devmem -- only the Tx test fails, may be related to our complicated
  .ndo_start_xmit() logic in ice, I'll investigate this as soon as
  I get some details from the val.

Thanks,
Olek

Re: [PATCH iwl-next v2 0/5] ice: add support for devmem/io_uring Rx and Tx

[Tony Nguyen]

On 1/29/2026 3:03 AM, Alexander Lobakin wrote:

...

> * io_uring tests pass, except for tests which try to setup RSS Rx
>    contexts -- after recent changes in the kernel, seems like at least
>    some of our drivers return -EOPNOTSUPP when trying to configure RSS.
> 
> Tony, did you have any related reports or planned fixes?

The only thing I'm aware of for ice RSS issues is for E830

ice: fix setting RSS VSI hash for E830
https://lore.kernel.org/intel-wired-lan/20260109085339.49839-1-marcin.szycik@linux.intel.com/

Thanks,
Tony

> * devmem -- only the Tx test fails, may be related to our complicated
>    .ndo_start_xmit() logic in ice, I'll investigate this as soon as
>    I get some details from the val.
> 
> Thanks,
> Olek

Re: [PATCH iwl-next v2 0/5] ice: add support for devmem/io_uring Rx and Tx

[Alexander Lobakin]

From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 29 Jan 2026 12:03:28 +0100

> From: Alexander Lobakin <aleksander.lobakin@intel.com>
> Date: Thu, 4 Dec 2025 16:51:28 +0100
> 
>> Now that ice uses libeth for managing Rx buffers and supports
>> configurable header split, it's ready to get support for sending
>> and receiving packets with unreadable (to the kernel) frags.
>>
>> Extend libeth just a little bit to allow creating PPs with custom
>> memory providers and make sure ice works correctly with the netdev
>> ops locking. Then add the full set of queue_mgmt_ops and don't
>> unmap unreadable frags on Tx completion.
>> No perf regressions for the regular flows and no code duplication
>> implied.
>>
>> Credits to the fbnic developers, which's code helped me understand
>> the memory providers and queue_mgmt_ops logics and served as
>> a reference.
>>
>> Alexander Lobakin (5):
>>   libeth: pass Rx queue index to PP when creating a fill queue
>>   libeth: handle creating pools with unreadable buffers
>>   ice: migrate to netdev ops lock
>>   ice: implement Rx queue management ops
>>   ice: add support for transmitting unreadable frags
>>
>>  drivers/net/ethernet/intel/ice/ice_lib.h    |  11 +-
>>  drivers/net/ethernet/intel/ice/ice_txrx.h   |   2 +
>>  drivers/net/ethernet/intel/idpf/idpf_txrx.h |   2 +
>>  include/net/libeth/rx.h                     |   2 +
>>  include/net/libeth/tx.h                     |   2 +-
>>  drivers/net/ethernet/intel/iavf/iavf_txrx.c |   1 +
>>  drivers/net/ethernet/intel/ice/ice_base.c   | 194 ++++++++++++++------
>>  drivers/net/ethernet/intel/ice/ice_lib.c    |  56 +++++-
>>  drivers/net/ethernet/intel/ice/ice_main.c   |  50 ++---
>>  drivers/net/ethernet/intel/ice/ice_sf_eth.c |   2 +
>>  drivers/net/ethernet/intel/ice/ice_txrx.c   |  43 +++--
>>  drivers/net/ethernet/intel/ice/ice_xsk.c    |   4 +-
>>  drivers/net/ethernet/intel/idpf/idpf_txrx.c |  13 ++
>>  drivers/net/ethernet/intel/libeth/rx.c      |  46 +++++
>>  14 files changed, 325 insertions(+), 103 deletions(-)
>>
>> ---
>> From v1[0]:
>> * rebase on top of the latest next-queue;
>> * fix a typo 'rxq_ixd' -> 'rxq_idx' (Tony).
>>
>> Testing hints:
>> * regular Rx and Tx for regressions;
>> * <kernel root>/tools/testing/selftests/drivers/net/hw/ contains
>>   scripts for testing netmem Rx and Tx, namely devmem.py and
>>   iou-zcrx.py (read the documentation first).
>>
>> [0] https://lore.kernel.org/intel-wired-lan/20251125173603.3834486-1-aleksander.lobakin@intel.com
> 
> Current status from the validation:
> 
> * io_uring tests pass, except for tests which try to setup RSS Rx
>   contexts -- after recent changes in the kernel, seems like at least
>   some of our drivers return -EOPNOTSUPP when trying to configure RSS.

Ok this might be due to that those scripts try to create new RSS
contexts. In fact zcrx works.

> 
> Tony, did you have any related reports or planned fixes?
> 
> * devmem -- only the Tx test fails, may be related to our complicated
>   .ndo_start_xmit() logic in ice, I'll investigate this as soon as
>   I get some details from the val.

In fact it works, it's just that by default without isolating TCP Rx
flows the completion packet may finish on any queue, not only the ones
bound to devmem. So ncdevmem reports "Ok" when the completion actually
hit the devmem Rx queue. Actual devmem Tx always works -- the output is
visible on the receiver side even when ncdevmem says "missing Tx
completion".

With that said, the only issue is that the bugfix was sent today for.
I'll merge it into the series and respin it once the window opens.

Thanks,
Olek