- 20 Dec, 2021 12 commits
-
-
Horatiu Vultur authored
The function lan966x_port_change_rx_flags() was used only when IFF_PROMISC flag was set. In that case it was setting to copy all the frames to the CPU instead of removing any RX filters. Therefore remove it. Fixes: d28d6d2e ("net: lan966x: add port module support") Signed-off-by:
Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Horatiu Vultur authored
This patch adds support for adding/removing mac entries in the SW list of entries and in the HW table. This is used by the bridge functionality. Signed-off-by:
-
Horatiu Vultur authored
This patch adds support for handling the interrupts generated by the analyzer. Currently, only the MAC table generates these interrupts. The MAC table will generate an interrupt whenever it learns or forgets an entry in the table. It is the SW responsibility figure out which entries were added/removed. Signed-off-by:
-
Horatiu Vultur authored
Extend dt-bindings for lan966x with analyzer interrupt. This interrupt can be generated for example when the HW learn/forgets an entry in the MAC table. Acked-by:
Rob Herring <robh@kernel.org> Reviewed-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by:
-
Horatiu Vultur authored
This patch adds the registers that will be used to enable switchdev and vlan functionality in the HW. Signed-off-by:
-
David S. Miller authored
Ido Schimmel says: ==================== mlxsw: devlink health reporter extensions This patchset extends the devlink health reporter registered by mlxsw to report new health events and their related parameters. These are meant to aid in debugging hardware and firmware issues. Patches #1-#2 are preparations. Patch #3 adds the definitions of the new events and parameters. Patch #4 extends the health reporter to report the new events and parameters. ==================== Signed-off-by: -
Danielle Ratson authored
Extend the devlink health reporter registered by mlxsw to report new health events and their related parameters. These are meant to aid in debugging of hardware / firmware issues. Beside the test event ('MLXSW_REG_MFDE_EVENT_ID_TEST') that is triggered following the devlink health 'test' sub-command, the new events are used to report the triggering of asserts in firmware code ('MLXSW_REG_MFDE_EVENT_ID_FW_ASSERT') and hardware issues ('MLXSW_REG_MFDE_EVENT_ID_FATAL_CAUSE'). Each event is accompanied with a severity parameter and per-event parameters that are meant to help root cause the detected issue. Signed-off-by:Danielle Ratson <danieller@nvidia.com> Signed-off-by:
Ido Schimmel <idosch@nvidia.com> Signed-off-by:
-
Danielle Ratson authored
Extend the Monitoring Firmware Debug (MFDE) register with new events and their related parameters. These events will be utilized by devlink-health in the next patch. Signed-off-by:
-
Danielle Ratson authored
Convert a series of if statements that handle different events to a switch case statement. Encapsulate the per-event code in different functions to simplify the code. This is a preparation for subsequent patches that will add more events that need to be handled. Signed-off-by:
-
Danielle Ratson authored
Currently, the MFDE register field names are using the convention: reg_mfde_<NAME_OF_FIELD>, and do not consider the name of the MFDE event. Fix the field names so they fit the more accurate convention: reg_mfde_<NAME_OF_EVENT>_<NAME_OF_FIELD>. Signed-off-by:
-
Manish Chopra authored
Commit 0a6890b9 ("bnx2x: Utilize FW 7.13.15.0.") added validation for fastpath HSI versions for different client init which was not meant for SR-IOV VF clients, which resulted in firmware asserts when running VF clients with different fastpath HSI version. This patch along with the new firmware support in patch #1 fixes this behavior in order to not validate fastpath HSI version for the VFs. Fixes: 0a6890b9 ("bnx2x: Utilize FW 7.13.15.0.") Signed-off-by:
Manish Chopra <manishc@marvell.com> Signed-off-by:
Prabhakar Kushwaha <pkushwaha@marvell.com> Signed-off-by:
Alok Prasad <palok@marvell.com> Signed-off-by:
Ariel Elior <aelior@marvell.com> Signed-off-by:
-
Manish Chopra authored
This new firmware addresses few important issues and enhancements as mentioned below - - Support direct invalidation of FP HSI Ver per function ID, required for invalidating FP HSI Ver prior to each VF start, as there is no VF start - BRB hardware block parity error detection support for the driver - Fix the FCOE underrun flow - Fix PSOD during FCoE BFS over the NIC ports after preboot driver - Maintains backward compatibility This patch incorporates this new firmware 7.13.21.0 in bnx2x driver. Signed-off-by:
-
- 19 Dec, 2021 14 commits
-
-
David S. Miller authored
Simon Horman says: ==================== allow user to offload tc action to net device Baowen Zheng says: Allow use of flow_indr_dev_register/flow_indr_dev_setup_offload to offload tc actions independent of flows. The motivation for this work is to prepare for using TC police action instances to provide hardware offload of OVS metering feature - which calls for policers that may be used by multiple flows and whose lifecycle is independent of any flows that use them. This patch includes basic changes to offload drivers to return EOPNOTSUPP if this feature is used - it is not yet supported by any driver. Tc cli command to offload and quote an action: # tc qdisc del dev $DEV ingress && sleep 1 || true # tc actions delete action police index 200 || true # tc qdisc add dev $DEV ingress # tc qdisc show dev $DEV ingress # tc actions add action police rate 100mbit burst 10000k index 200 skip_sw # tc -s -d actions list action police total acts 1 action order 0: police 0xc8 rate 100Mbit burst 10000Kb mtu 2Kb action reclassify overhead 0b linklayer ethernet ref 1 bind 0 installed 142 sec used 0 sec Action statistics: Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 skip_sw in_hw in_hw_count 1 used_hw_stats delayed # tc filter add dev $DEV protocol ip parent ffff: \ flower skip_sw ip_proto tcp action police index 200 # tc -s -d filter show dev $DEV protocol ip parent ffff: filter pref 49152 flower chain 0 filter pref 49152 flower chain 0 handle 0x1 eth_type ipv4 ip_proto tcp skip_sw in_hw in_hw_count 1 action order 1: police 0xc8 rate 100Mbit burst 10000Kb mtu 2Kb action reclassify overhead 0b linklayer ethernet ref 2 bind 1 installed 300 sec used 0 sec Action statistics: Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 skip_sw in_hw in_hw_count 1 used_hw_stats delayed # tc filter add dev $DEV protocol ipv6 parent ffff: \ flower skip_sw ip_proto tcp action police index 200 # tc -s -d filter show dev $DEV protocol ipv6 parent ffff: filter pref 49151 flower chain 0 filter pref 49151 flower chain 0 handle 0x1 eth_type ipv6 ip_proto tcp skip_sw in_hw in_hw_count 1 action order 1: police 0xc8 rate 100Mbit burst 10000Kb mtu 2Kb action reclassify overhead 0b linklayer ethernet ref 3 bind 2 installed 761 sec used 0 sec Action statistics: Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 skip_sw in_hw in_hw_count 1 used_hw_stats delayed # tc -s -d actions list action police total acts 1 action order 0: police 0xc8 rate 100Mbit burst 10000Kb mtu 2Kb action reclassify overhead 0b linklayer ethernet ref 3 bind 2 installed 917 sec used 0 sec Action statistics: Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 skip_sw in_hw in_hw_count 1 used_hw_stats delayed Changes between v7 and v8: * Rename enum offload_act_command as enum flow_offload_act_command * Add a single patch to rename exts stats update function for readability. * Fix the kernel test robot issue reported by Oliver Sang. Changes between v6 and v7: * Add a single patch to rename offload functions with offload for readability. * Post 166b6a46 ("flow_offload: return EOPNOTSUPP for the unsupported mpls action type") as a bug fix to netdev. * Rename enum flow_act_command as enum flow_offload_act_command * Rename the new added action offload setup ops with offload instead of flow * Rename the new added action offload function with offload instead of flow. * Add more selftest cases for validate filter and actions. Changes between v5 and v6: * Fix issue reported by Dan Carpenter found using Smatch. Changes beteeen v4 and v5: * Made changes of code style according to the public review comments. * Add a fix for unsupported mpls action type in flow action setup stage. * Add ops to tc_action_ops for flow action setup to facilitate adding a standalone action module. * Add notification process when deleting action in reoffload process. Changes between v3 and v4: * Made changes according to the public review comments. * Validate flags inside tcf_action_init() instead of creating new tcf_exts_validate_actions() function. * Exactly match when validating flags of actions and filters. * Add index to flow_action_entry for driver to identify actions. Changes between v2 and v3: * Made changes according to the review comments. * Delete in_hw and not_in_hw flag and user can judge if the action is offloaded to any hardware by in_hw_count. * Split the main patch of the action offload to three single patch to facilitate code review. Changes between v1 and v2: * Add the skip_hw/skip_sw for user to specify if the action should be in hardware or software. * Fix issue of sleeping function called from invalid context. * Change the action offload/delete from batch to one by one. * Add some parameters to the netlink message for user space to look up the offload status of the actions. * Add reoffload process to update action hw_count when driver is inserted or removed. Changes between v1 and RFC: * Fix robot test failure. * Change actions offload process in action add function rather than action init. * Change actions offload delete process after tcf_del_notify to keep undeleted actions. * Add process to update actions stats from hardware. ==================== Signed-off-by: -
Baowen Zheng authored
Add selftest cases in action police with skip_hw. Add selftest case to validate flags of filter and action. These tests depend on corresponding iproute2 command support. Signed-off-by:
Baowen Zheng <baowen.zheng@corigine.com> Signed-off-by:
Simon Horman <simon.horman@corigine.com> Signed-off-by:
-
Baowen Zheng authored
Add process to validate flags of filter and actions when adding a tc filter. We need to prevent adding filter with flags conflicts with its actions. Signed-off-by:
Louis Peens <louis.peens@corigine.com> Signed-off-by:
Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
Baowen Zheng authored
Add reoffload process to update hw_count when driver is inserted or removed. We will delete the action if it is with skip_sw flag and not offloaded to any hardware in reoffload process. When reoffloading actions, we still offload the actions that are added independent of filters. Signed-off-by:
-
Baowen Zheng authored
Save full action flags and return user flags when return flags to user space. Save full action flags to distinguish if the action is created independent from classifier. We made this change mainly for further patch to reoffload tc actions. Signed-off-by:
-
Baowen Zheng authored
When collecting stats for actions update them using both hardware and software counters. Stats update process should not run in context of preempt_disable. Signed-off-by:
-
Baowen Zheng authored
Rename exts stats update functions with hw for readability. We make this change also to update stats from hw for an action when it is offloaded to hw as a single action. Signed-off-by:
-
Baowen Zheng authored
We add skip_hw and skip_sw for user to control if offload the action to hardware. We also add in_hw_count for user to indicate if the action is offloaded to any hardware. Signed-off-by:
-
Baowen Zheng authored
Use flow_indr_dev_register/flow_indr_dev_setup_offload to offload tc action. We need to call tc_cleanup_flow_action to clean up tc action entry since in tc_setup_action, some actions may hold dev refcnt, especially the mirror action. Signed-off-by:
-
Baowen Zheng authored
Add a new ops to tc_action_ops for flow action setup. Refactor function tc_setup_flow_action to use this new ops. We make this change to facilitate to add standalone action module. We will also use this ops to offload action independent of filter in following patch. Signed-off-by:
-
Baowen Zheng authored
To improves readability, we rename offload functions with offload instead of flow. The term flow is related to exact matches, so we rename these functions with offload. We make this change to facilitate single action offload functions naming. Signed-off-by:
-
Baowen Zheng authored
Add index to flow_action_entry structure and delete index from police and gate child structure. We make this change to offload tc action for driver to identify a tc action. Signed-off-by:
-
Baowen Zheng authored
A follow-up patch will allow users to offload tc actions independent of classifier in the software datapath. In preparation for this, teach all drivers that support offload of the flow tables to reject such configuration as currently none of them support it. Signed-off-by:
-
Baowen Zheng authored
Fill flags to action structure to allow user control if the action should be offloaded to hardware or not. Signed-off-by:
-
- 18 Dec, 2021 9 commits
-
-
Yajun Deng authored
The xdp_rxq_info_unreg() called by xdp_rxq_info_reg() is meaningless when dev is NULL, so move the if dev statements to the first. Signed-off-by:
Yajun Deng <yajun.deng@linux.dev> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queueDavid S. Miller authored
Tony Nguyen says: ==================== 40GbE Intel Wired LAN Driver Updates 2021-12-17 Brett Creeley says: This patch series adds support in the iavf driver for communicating and using VIRTCHNL_VF_OFFLOAD_VLAN_V2. The current VIRTCHNL_VF_OFFLOAD_VLAN is very limited and covers all 802.1Q VLAN offloads and filtering with no granularity. The new VIRTCHNL_VF_OFFLOAD_VLAN_V2 adds more granularity, flexibility, and support for 802.1ad offloads and filtering. This includes the VF negotiating which VLAN offloads/filtering it's allowed, where VLAN tags should be inserted and/or stripped into and from descriptors, and the supported VLAN protocols. ==================== Signed-off-by: -
Jakub Kicinski authored
Mat Martineau says: ==================== mptcp: Miscellaneous changes for 5.17 These are three unrelated patches that we've been testing in the MPTCP tree. Patch 1 modifies the packet scheduler that picks which TCP subflow is used for each chunk of outgoing data. The updated scheduler improves throughput on multiple-subflow connections. Patch 2 updates a selftest to verify recent TCP_ULP sockopt changes on MPTCP fallback sockets. Patch 3 cleans up some unnecessary comparisons with an 8-bit value. ==================== Link: https://lore.kernel.org/r/20211217233702.299461-1-mathew.j.martineau@linux.intel.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jean Sacren authored
entry->addr.id is u8 with a range from 0 to 255 and MAX_ADDR_ID is 255. We should drop both false expressions of (entry->addr.id > MAX_ADDR_ID). We should also remove the obsolete parentheses in the first if branch. Use U8_MAX for MAX_ADDR_ID and add a comment to show the link to mptcp_addr_info.id as suggested by Mr. Matthieu Baerts. Reviewed-by:
Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by:
Jean Sacren <sakiwit@gmail.com> Signed-off-by:
Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by:
-
Florian Westphal authored
The kernel will crash without 'mptcp: clear 'kern' flag from fallback sockets' change. Since this doesn't slow down testing in a noticeable way, run this unconditionally. The explicit test did not catch this, because the check was done for tcp socket returned by 'socket(.. IPPROTO_TCP) rather than a tcp socket returned by accept() on a mptcp listen fd. Signed-off-by:
Florian Westphal <fw@strlen.de> Signed-off-by:
-
Paolo Abeni authored
The MPTCP packet scheduler has sub-optimal behavior with asymmetric subflows: if the faster subflow-level cwin is closed, the packet scheduler can enqueue "too much" data on a slower subflow. When all the data on the faster subflow is acked, if the mptcp-level cwin is closed, and link utilization becomes suboptimal. The solution is implementing blest-like[1] HoL-blocking estimation, transmitting only on the subflow with the shorter estimated time to flush the queued memory. If such subflows cwin is closed, we wait even if other subflows are available. This is quite simpler than the original blest implementation, as we leverage the pacing rate provided by the TCP socket. To get a more accurate estimation for the subflow linger-time, we maintain a per-subflow weighted average of such info. Additionally drop magic numbers usage in favor of newly defined macros and use more meaningful names for status variable. [1] http://dl.ifip.org/db/conf/networking/networking2016/1570234725.pdf Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/137Reviewed-by:
Paolo Abeni <pabeni@redhat.com> Signed-off-by:
-
Dan Carpenter authored
The debugfs_create_dir() function never returns NULL. It does return error pointers but in normal situations like this there is no need to check for errors. Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Link: https://lore.kernel.org/r/20211217071037.GE26548@kiliSigned-off-by:
-
Dan Carpenter authored
There is already an "int err" declared at the start of the function so re-use that instead of declaring a shadow err variable. Signed-off-by:
-
Gerhard Engleder authored
The following warning is fixed with additional config dependencies: s390-linux-ld: drivers/net/ethernet/engleder/tsnep_main.o: in function `tsnep_probe': tsnep_main.c:(.text+0x1de6): undefined reference to `devm_ioremap_resource' Reported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
Gerhard Engleder <gerhard@engleder-embedded.com> Link: https://lore.kernel.org/r/20211216200154.1520-1-gerhard@engleder-embedded.comSigned-off-by:
-
- 17 Dec, 2021 5 commits
-
-
Brett Creeley authored
For VIRTCHNL_VF_OFFLOAD_VLAN, PF's would limit the number of VLAN filters a VF was allowed to add. However, by the time the opcode failed, the VLAN netdev had already been added. VIRTCHNL_VF_OFFLOAD_VLAN_V2 added the ability for a PF to tell the VF how many VLAN filters it's allowed to add. Make changes to support that functionality. Signed-off-by:
Brett Creeley <brett.creeley@intel.com> Tested-by:
Konrad Jankowski <konrad0.jankowski@intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com>
-
Brett Creeley authored
The new VIRTCHNL_VF_OFFLOAD_VLAN_V2 capability added support that allows the VF to support 802.1Q and 802.1ad VLAN insertion and stripping if successfully negotiated via VIRTCHNL_OP_GET_OFFLOAD_VLAN_V2_CAPS. Multiple changes were needed to support this new functionality. 1. Added new aq_required flags to support any kind of VLAN stripping and insertion offload requests via virtchnl. 2. Added the new method iavf_set_vlan_offload_features() that's used during VF initialization, VF reset, and iavf_set_features() to set the aq_required bits based on the current VLAN offload configuration of the VF's netdev. 3. Added virtchnl handling for VIRTCHNL_OP_ENABLE_STRIPPING_V2, VIRTCHNL_OP_DISABLE_STRIPPING_V2, VIRTCHNL_OP_ENABLE_INSERTION_V2, and VIRTCHNL_OP_ENABLE_INSERTION_V2. Signed-off-by:
-
Brett Creeley authored
The new VIRTCHNL_VF_OFFLOAD_VLAN_V2 capability added support that allows the PF to set the location of the Tx and Rx VLAN tag for insertion and stripping offloads. In order to support this functionality a few changes are needed. 1. Add a new method to cache the VLAN tag location based on negotiated capabilities for the Tx and Rx ring flags. This needs to be called in the initialization and reset paths. 2. Refactor the transmit hotpath to account for the new Tx ring flags. When IAVF_TXR_FLAGS_VLAN_LOC_L2TAG2 is set, then the driver needs to insert the VLAN tag in the L2TAG2 field of the transmit descriptor. When the IAVF_TXRX_FLAGS_VLAN_LOC_L2TAG1 is set, then the driver needs to use the l2tag1 field of the data descriptor (same behavior as before). 3. Refactor the iavf_tx_prepare_vlan_flags() function to simplify transmit hardware VLAN offload functionality by only depending on the skb_vlan_tag_present() function. This can be done because the OS won't request transmit offload for a VLAN unless the driver told the OS it's supported and enabled. 4. Refactor the receive hotpath to account for the new Rx ring flags and VLAN ethertypes. This requires checking the Rx ring flags and descriptor status bits to determine the location of the VLAN tag. Also, since only a single ethertype can be supported at a time, check the enabled netdev features before specifying a VLAN ethertype in __vlan_hwaccel_put_tag(). Signed-off-by:
-
Brett Creeley authored
Based on VIRTCHNL_VF_OFFLOAD_VLAN_V2, the VF can now support more VLAN capabilities (i.e. 802.1AD offloads and filtering). In order to communicate these capabilities to the netdev layer, the VF needs to parse its VLAN capabilities based on whether it was able to negotiation VIRTCHNL_VF_OFFLOAD_VLAN or VIRTCHNL_VF_OFFLOAD_VLAN_V2 or neither of these. In order to support this, add the following functionality: iavf_get_netdev_vlan_hw_features() - This is used to determine the VLAN features that the underlying hardware supports and that can be toggled off/on based on the negotiated capabiltiies. For example, if VIRTCHNL_VF_OFFLOAD_VLAN_V2 was negotiated, then any capability marked with VIRTCHNL_VLAN_TOGGLE can be toggled on/off by the VF. If VIRTCHNL_VF_OFFLOAD_VLAN was negotiated, then only VLAN insertion and/or stripping can be toggled on/off. iavf_get_netdev_vlan_features() - This is used to determine the VLAN features that the underlying hardware supports and that should be enabled by default. For example, if VIRTHCNL_VF_OFFLOAD_VLAN_V2 was negotiated, then any supported capability that has its ethertype_init filed set should be enabled by default. If VIRTCHNL_VF_OFFLOAD_VLAN was negotiated, then filtering, stripping, and insertion should be enabled by default. Also, refactor iavf_fix_features() to take into account the new capabilities. To do this, query all the supported features (enabled by default and toggleable) and make sure the requested change is supported. If VIRTCHNL_VF_OFFLOAD_VLAN_V2 is successfully negotiated, there is no need to check VIRTCHNL_VLAN_TOGGLE here because the driver already told the netdev layer which features can be toggled via netdev->hw_features during iavf_process_config(), so only those features will be requested to change. Signed-off-by:
-
Brett Creeley authored
In order to support the new VIRTCHNL_VF_OFFLOAD_VLAN_V2 capability the VF driver needs to rework it's initialization state machine and reset flow. This has to be done because successful negotiation of VIRTCHNL_VF_OFFLOAD_VLAN_V2 requires the VF driver to perform a second capability request via VIRTCHNL_OP_GET_OFFLOAD_VLAN_V2_CAPS before configuring the adapter and its netdev. Add the VIRTCHNL_VF_OFFLOAD_VLAN_V2 bit when sending the VIRTHCNL_OP_GET_VF_RESOURECES message. The underlying PF will either support VIRTCHNL_VF_OFFLOAD_VLAN or VIRTCHNL_VF_OFFLOAD_VLAN_V2 or neither. Both of these offloads should never be supported together. Based on this, add 2 new states to the initialization state machine: __IAVF_INIT_GET_OFFLOAD_VLAN_V2_CAPS __IAVF_INIT_CONFIG_ADAPTER The __IAVF_INIT_GET_OFFLOAD_VLAN_V2_CAPS state is used to request/store the new VLAN capabilities if and only if VIRTCHNL_VLAN_OFFLOAD_VLAN_V2 was successfully negotiated in the __IAVF_INIT_GET_RESOURCES state. The __IAVF_INIT_CONFIG_ADAPTER state is used to configure the adapter/netdev after the resource requests have finished. The VF will move into this state regardless of whether it successfully negotiated VIRTCHNL_VF_OFFLOAD_VLAN or VIRTCHNL_VF_OFFLOAD_VLAN_V2. Also, add a the new flag IAVF_FLAG_AQ_GET_OFFLOAD_VLAN_V2_CAPS and set it during VF reset. If VIRTCHNL_VF_OFFLOAD_VLAN_V2 was successfully negotiated then the VF will request its VLAN capabilities via VIRTCHNL_OP_GET_OFFLOAD_VLAN_V2_CAPS during the reset. This is needed because the PF may change/modify the VF's configuration during VF reset (i.e. modifying the VF's port VLAN configuration). This also, required the VF to call netdev_update_features() since its VLAN features may change during VF reset. Make sure to call this under rtnl_lock(). Signed-off-by:
-
