Changes in netc_switch.h

05b5ee610fbb8ca4ce9dc21299442aa827b38008 - net: dsa: netc: implement dynamic FDB entry ageing

Thu, 11 Jun 2026 04:14:58 +0200

net: dsa: netc: implement dynamic FDB entry ageingThe NETC switch does not age out dynamic FDB entries automatically.Without software management, stale entries persist after topologychanges and cause incorrect forwarding.Add a delayed work that periodically removes entries that have not beenrefreshed within the specified cycles. The effective ageing time is: ageing_time = fdbt_ageing_delay * 100Default values are 3s interval and 100 cycles (300s total), matchingthe IEEE 802.1Q default ageing time. The work starts when the firstport joins a bridge (tracked via br_cnt) and is cancelled when thelast port leaves. All FDB operations are serialized under fdbt_lock.Implement .set_ageing_time() to allow the bridge layer to reconfigureageing parameters on demand.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260611021458.2629145-10-wei.fang@oss.nxp.comSigned-off-by: Jakub Kicinski List of files: /linux/drivers/net/dsa/netc/netc_switch.h

751aa5a5d593e19204d3e8ed0787a18a29a18690 - net: dsa: netc: add bridge mode support

Thu, 11 Jun 2026 04:14:57 +0200

net: dsa: netc: add bridge mode supportWire up the port_bridge_join, port_bridge_leave and port_vlan_filteringDSA callbacks to support both VLAN-unaware and VLAN-aware bridge modes.For VLAN-unaware bridges, each bridge instance is assigned a dedicatedinternal PVID via NETC_VLAN_UNAWARE_PVID(bridge.num), counting downfrom VID 4095. A VFT entry is created for this PVID with hardware MAClearning and flood-on-miss forwarding enabled. The CPU port is includedas a VFT member so frames can reach the host. The reserved VID range isblocked in port_vlan_add to prevent user-space conflicts.Only one VLAN-aware bridge is supported at a time; this constraint isenforced in port_bridge_join and port_vlan_filtering. The per-port PVIDis tracked in software and written to the BPDVR register whenever VLANfiltering is active.When a port leaves the bridge, its dynamic FDB entries are flushed rightaway in port_bridge_leave(), without waiting for the ageing cycle. Whena link down event occurs on a port, netc_mac_link_down() will also clearthe port's dynamic FDB entries via netc_port_remove_dynamic_entries().Non-bridge ports have no dynamic FDB entries, so this call is alwayssafe. Additionally, .port_fast_age() callback is added to flush thedynamic FDB entries associated to a port.Host flood rules are removed from the ingress port filter table when aport joins a bridge to avoid bypassing FDB lookup and MAC learning.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260611021458.2629145-9-wei.fang@oss.nxp.comSigned-off-by: Jakub Kicinski List of files: /linux/drivers/net/dsa/netc/netc_switch.h

84b4a3b30abd226ec528985d7c3cb63a3376d098 - net: dsa: netc: add VLAN filter table and egress treatment management

Thu, 11 Jun 2026 04:14:56 +0200

net: dsa: netc: add VLAN filter table and egress treatment managementImplement the DSA .port_vlan_add and .port_vlan_del operations to enableVLAN-aware bridge offloading on the NETC switch.VLAN membership is maintained in the VLAN Filter Table (VFT). Adding thefirst port to a VLAN creates a new VFT entry with hardware MAC learningand flood-on-miss forwarding; subsequent ports update the existingentry's membership bitmap. Removing the last port deletes the entry.Egress tagging is handled through the Egress Treatment Table (ETT). EachVLAN is allocated a group of ETT entries, one per available port. Portsare assigned a sequential ett_offset during initialisation, used toaddress each port's entry within the group. Untagged ports configure theETT to strip the outer VLAN tag; tagged ports pass frames throughunmodified. Each ETT group is optionally paired with an Egress CounterTable (ECT) group for per-port frame counting, allocated on a best-effortbasis. When the egress rule of an ETT entry changes, the counter of thecorresponding ECT entry will be recounted to track the number of framesthat match the new egress rule.A software shadow list serialised by vft_lock tracks active VLAN stateacross both port membership and egress tagging. VID 0 is used for singleport mode and is ignored by both callbacks.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260611021458.2629145-8-wei.fang@oss.nxp.comSigned-off-by: Jakub Kicinski List of files: /linux/drivers/net/dsa/netc/netc_switch.h

beb0e54f3806cf01a24740b23ec01c4fab186b5c - net: dsa: netc: add support for ethtool private statistics

Mon, 18 May 2026 10:25:06 +0200

net: dsa: netc: add support for ethtool private statisticsImplement the ethtool private statistics interface to expose additionalport-level and MAC-level counters that are not covered by the standardIEEE 802.3 statistics. The pMAC counters are only reported when the portsupports Frame Preemption (802.1Qbu/802.3br).Note that although rtnl_link_stats64 provides some standard statisticssuch as rx octets, rx frame errors, rx dropped packets, and tx packets,these are overall port statistics. The NETC switch supports preemptionon each port, and each port has two MACs (eMAC and pMAC). The driverprivate statistics are used to obtain statistics for each MAC, allowingusers to perform analysis and debugging.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260518082506.1318236-16-wei.fang@nxp.comSigned-off-by: Paolo Abeni List of files: /linux/drivers/net/dsa/netc/netc_switch.h

25049d8b6e6b87f7ffcf53ce5ea1b51528b8677f - net: dsa: netc: add support for the standardized counters

Mon, 18 May 2026 10:25:05 +0200

net: dsa: netc: add support for the standardized countersEach user port of the NETC switch supports 802.3 basic and mandatorymanaged objects statistic counters and IETF Management InformationDatabase (MIB) package (RFC2665) and Remote Network Monitoring (RMON)counters. And all of these counters are 64-bit registers. In addition,some user ports support preemption, so these ports have two MACs, MAC0 is the express MAC (eMAC), MAC 1 is the preemptible MAC (pMAC). Sofor ports that support preemption, the statistics are the sum of thepMAC and eMAC statistics.Note that the current switch driver does not support preemption, allframes are sent and received via the eMAC by default. The statisticsread from the pMAC should be zero.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260518082506.1318236-15-wei.fang@nxp.comSigned-off-by: Paolo Abeni List of files: /linux/drivers/net/dsa/netc/netc_switch.h

a5ccb7f5e067eae23707a61e4fc5d6214b0e4777 - net: dsa: netc: initialize buffer pool table and implement flow-control

Mon, 18 May 2026 10:25:04 +0200

net: dsa: netc: initialize buffer pool table and implement flow-controlThe buffer pool is a quantity of memory available for buffering a groupof flows (e.g. frames having the same priority, frames received from thesame port), while waiting to be transmitted on a port. The buffer pooltracks internal memory consumption with upper bound limits and optionallya non-shared portion when associated with a shared buffer pool. Currentlythe shared buffer pool is not supported, it will be added in the future.For i.MX94, the switch has 4 ports and 8 buffer pools, so each port isallocated two buffer pools. For frames with priorities of 0 to 3, theywill be mapped to the first buffer pool; For frames with priorities of4 to 7, they will be mapped to the second buffer pool. Each buffer poolhas a flow control on threshold and a flow control off threshold. Bysetting these threshold, add the flow control support to each port.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260518082506.1318236-14-wei.fang@nxp.comSigned-off-by: Paolo Abeni List of files: /linux/drivers/net/dsa/netc/netc_switch.h

46d6407692c80b258ecba3af831130a6f6e2feea - net: dsa: netc: add FDB, STP, MTU, port setup and host flooding support

Mon, 18 May 2026 10:25:03 +0200

net: dsa: netc: add FDB, STP, MTU, port setup and host flooding supportExpand the NETC switch driver with several foundational features:- FDB and MDB management- STP state handling- MTU configuration- Port setup/teardown- Host flooding supportAt this stage, the driver operates only in standalone port mode. Eachport uses VLAN 0 as its PVID, meaning ingress frames are internallyassigned VID 0 regardless of whether they arrive tagged or untagged.Note that this does not inject a VLAN 0 header into the frame, the VIDis used purely for subsequent VLAN processing within the switch.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260518082506.1318236-13-wei.fang@nxp.comSigned-off-by: Paolo Abeni List of files: /linux/drivers/net/dsa/netc/netc_switch.h

bbe97e34721984b5d58ad343f0e6c3441d9c8c4a - net: dsa: netc: add phylink MAC operations

Mon, 18 May 2026 10:25:02 +0200

net: dsa: netc: add phylink MAC operationsDifferent versions of NETC switches have different numbers of ports andMAC capabilities. Add .phylink_get_caps() to struct netc_switch_info,allowing each NETC switch version to implement its own callback forobtaining MAC capabilities.Implement the phylink_mac_ops callbacks: .mac_config(), .mac_link_up(),and .mac_link_down(). Note that flow-control configuration is not yetsupported in .mac_link_up(), but will be implemented in a subsequentpatch.Signed-off-by: Wei Fang Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260518082506.1318236-12-wei.fang@nxp.comSigned-off-by: Paolo Abeni List of files: /linux/drivers/net/dsa/netc/netc_switch.h

187fbae024c8439533d7f075f4ab4b594dc19992 - net: dsa: netc: introduce NXP NETC switch driver for i.MX94

Mon, 18 May 2026 10:25:01 +0200

net: dsa: netc: introduce NXP NETC switch driver for i.MX94For i.MX94 series, the NETC IP provides full 802.1Q Ethernet switchfunctionality, advanced QoS with 8 traffic classes, and a full range ofTSN standards capabilities. The switch has 3 user ports and 1 CPU port,the CPU port is connected to an internal ENETC. Since the switch and theinternal ENETC are fully integrated within the NETC IP, no back-to-backMAC connection is required. Instead, a light-weight "pseudo MAC" is usedbetween the switch and the ENETC. This translates to lower power (lesslogic and memory) and lower delay (as there is no serialization delayacross this link).Introduce the initial NETC switch driver with basic probe and removefunctionality. More features will be added in subsequent patches.Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260518082506.1318236-11-wei.fang@nxp.comSigned-off-by: Paolo Abeni List of files: /linux/drivers/net/dsa/netc/netc_switch.h