summaryrefslogtreecommitdiffstats
path: root/drivers/net
Commit message (Collapse)AuthorAgeFilesLines
* Merge tag 'net-6.4-rc1' of ↵Linus Torvalds2023-05-0546-312/+602
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net Pull networking fixes from Jakub Kicinski: "Including fixes from netfilter. Current release - regressions: - sched: act_pedit: free pedit keys on bail from offset check Current release - new code bugs: - pds_core: - Kconfig fixes (DEBUGFS and AUXILIARY_BUS) - fix mutex double unlock in error path Previous releases - regressions: - sched: cls_api: remove block_cb from driver_list before freeing - nf_tables: fix ct untracked match breakage - eth: mtk_eth_soc: drop generic vlan rx offload - sched: flower: fix error handler on replace Previous releases - always broken: - tcp: fix skb_copy_ubufs() vs BIG TCP - ipv6: fix skb hash for some RST packets - af_packet: don't send zero-byte data in packet_sendmsg_spkt() - rxrpc: timeout handling fixes after moving client call connection to the I/O thread - ixgbe: fix panic during XDP_TX with > 64 CPUs - igc: RMW the SRRCTL register to prevent losing timestamp config - dsa: mt7530: fix corrupt frames using TRGMII on 40 MHz XTAL MT7621 - r8152: - fix flow control issue of RTL8156A - fix the poor throughput for 2.5G devices - move setting r8153b_rx_agg_chg_indicate() to fix coalescing - enable autosuspend - ncsi: clear Tx enable mode when handling a Config required AEN - octeontx2-pf: macsec: fixes for CN10KB ASIC rev Misc: - 9p: remove INET dependency" * tag 'net-6.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (69 commits) net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop() pds_core: fix mutex double unlock in error path net/sched: flower: fix error handler on replace Revert "net/sched: flower: Fix wrong handle assignment during filter change" net/sched: flower: fix filter idr initialization net: fec: correct the counting of XDP sent frames bonding: add xdp_features support net: enetc: check the index of the SFI rather than the handle sfc: Add back mailing list virtio_net: suppress cpu stall when free_unused_bufs ice: block LAN in case of VF to VF offload net: dsa: mt7530: fix network connectivity with multiple CPU ports net: dsa: mt7530: fix corrupt frames using trgmii on 40 MHz XTAL MT7621 9p: Remove INET dependency netfilter: nf_tables: fix ct untracked match breakage af_packet: Don't send zero-byte data in packet_sendmsg_spkt(). igc: read before write to SRRCTL register pds_core: add AUXILIARY_BUS and NET_DEVLINK to Kconfig pds_core: remove CONFIG_DEBUG_FS from makefile ionic: catch failure from devlink_alloc ...
| * Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextJakub Kicinski2023-05-053-2/+33
| |\ | | | | | | | | | | | | | | | | | | There's a fix which landed in net-next, pull it in along with the couple of minor cleanups. Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| | * bonding: add xdp_features supportLorenzo Bianconi2023-05-052-0/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce xdp_features support for bonding driver according to the slave devices attached to the master one. xdp_features is required whenever we want to xdp_redirect traffic into a bond device and then into selected slaves attached to it. Reviewed-by: Simon Horman <simon.horman@corigine.com> Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com> Fixes: 66c0e13ad236 ("drivers: net: turn on XDP features") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Reviewed-by: Jussi Maki <joamaki@gmail.com> Tested-by: Jussi Maki <joamaki@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| | * atlantic:hw_atl2:hw_atl2_utils_fw: Remove unnecessary (void*) conversionswuych2023-04-281-2/+2
| | | | | | | | | | | | | | | | | | | | | Pointer variables of void * type do not require type cast. Signed-off-by: wuych <yunchuan@nfschina.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop()Florian Fainelli2023-05-051-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The call to phy_stop() races with the later call to phy_disconnect(), resulting in concurrent phy_suspend() calls being run from different CPUs. The final call to phy_disconnect() ensures that the PHY is stopped and suspended, too. Fixes: c96e731c93ff ("net: bcmgenet: connect and disconnect from the PHY state machine") Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | pds_core: fix mutex double unlock in error pathShannon Nelson2023-05-051-8/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a double unlock in an error handling path by unlocking as soon as the error is seen and removing unlocks in the error cleanup path. Link: https://lore.kernel.org/kernel-janitors/209a09f6-5ec6-40c7-a5ec-6260d8f54d25@kili.mountain/ Fixes: 523847df1b37 ("pds_core: add devcmd device interfaces") Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Shannon Nelson <shannon.nelson@amd.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: fec: correct the counting of XDP sent framesShenwei Wang2023-05-051-4/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the current xdp_xmit implementation, if any single frame fails to transmit due to insufficient buffer descriptors, the function nevertheless reports success in sending all frames. This results in erroneously indicating that frames were transmitted when in fact they were dropped. This patch fixes the issue by ensureing the return value properly indicates the actual number of frames successfully transmitted, rather than potentially reporting success for all frames when some could not transmit. Fixes: 6d6b39f180b8 ("net: fec: add initial XDP support") Signed-off-by: Gagandeep Singh <g.singh@nxp.com> Signed-off-by: Shenwei Wang <shenwei.wang@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: enetc: check the index of the SFI rather than the handleWei Fang2023-05-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We should check whether the current SFI (Stream Filter Instance) table is full before creating a new SFI entry. However, the previous logic checks the handle by mistake and might lead to unpredictable behavior. Fixes: 888ae5a3952b ("net: enetc: add tc flower psfp offload driver") Signed-off-by: Wei Fang <wei.fang@nxp.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | virtio_net: suppress cpu stall when free_unused_bufsWenliang Wang2023-05-051-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For multi-queue and large ring-size use case, the following error occurred when free_unused_bufs: rcu: INFO: rcu_sched self-detected stall on CPU. Fixes: 986a4f4d452d ("virtio_net: multiqueue support") Signed-off-by: Wenliang Wang <wangwenliang.1995@bytedance.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | ice: block LAN in case of VF to VF offloadMichal Swiatkowski2023-05-051-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | VF to VF traffic shouldn't go outside. To enforce it, set only the loopback enable bit in case of all ingress type rules added via the tc tool. Fixes: 0d08a441fb1a ("ice: ndo_setup_tc implementation for PF") Reported-by: Sujai Buvaneswaran <Sujai.Buvaneswaran@intel.com> Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: George Kuruvinakunnel <george.kuruvinakunnel@intel.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: dsa: mt7530: fix network connectivity with multiple CPU portsArınç ÜNAL2023-05-051-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On mt753x_cpu_port_enable() there's code that enables flooding for the CPU port only. Since mt753x_cpu_port_enable() runs twice when both CPU ports are enabled, port 6 becomes the only port to forward the frames to. But port 5 is the active port, so no frames received from the user ports will be forwarded to port 5 which breaks network connectivity. Every bit of the BC_FFP, UNM_FFP, and UNU_FFP bits represents a port. Fix this issue by setting the bit that corresponds to the CPU port without overwriting the other bits. Clear the bits beforehand only for the MT7531 switch. According to the documents MT7621 Giga Switch Programming Guide v0.3 and MT7531 Reference Manual for Development Board v1.0, after reset, the BC_FFP, UNM_FFP, and UNU_FFP bits are set to 1 for MT7531, 0 for MT7530. The commit 5e5502e012b8 ("net: dsa: mt7530: fix roaming from DSA user ports") silently changed the method to set the bits on the MT7530_MFC. Instead of clearing the relevant bits before mt7530_cpu_port_enable() which runs under a for loop, the commit started doing it on mt7530_cpu_port_enable(). Back then, this didn't really matter as only a single CPU port could be used since the CPU port number was hardcoded. The driver was later changed with commit 1f9a6abecf53 ("net: dsa: mt7530: get cpu-port via dp->cpu_dp instead of constant") to retrieve the CPU port via dp->cpu_dp. With that, this silent change became an issue for when using multiple CPU ports. Fixes: 5e5502e012b8 ("net: dsa: mt7530: fix roaming from DSA user ports") Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: dsa: mt7530: fix corrupt frames using trgmii on 40 MHz XTAL MT7621Arınç ÜNAL2023-05-051-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The multi-chip module MT7530 switch with a 40 MHz oscillator on the MT7621AT, MT7621DAT, and MT7621ST SoCs forwards corrupt frames using trgmii. This is caused by the assumption that MT7621 SoCs have got 150 MHz PLL, hence using the ncpo1 value, 0x0780. My testing shows this value works on Unielec U7621-06, Bartel's testing shows it won't work on Hi-Link HLK-MT7621A and Netgear WAC104. All devices tested have got 40 MHz oscillators. Using the value for 125 MHz PLL, 0x0640, works on all boards at hand. The definitions for 125 MHz PLL exist on the Banana Pi BPI-R2 BSP source code whilst 150 MHz PLL don't. Forwarding frames using trgmii on the MCM MT7530 switch with a 25 MHz oscillator on the said MT7621 SoCs works fine because the ncpo1 value defined for it is for 125 MHz PLL. Change the 150 MHz PLL comment to 125 MHz PLL, and use the 125 MHz PLL ncpo1 values for both oscillator frequencies. Link: https://github.com/BPI-SINOVOIP/BPI-R2-bsp/blob/81d24bbce7d99524d0771a8bdb2d6663e4eb4faa/u-boot-mt/drivers/net/rt2880_eth.c#L2195 Fixes: 7ef6f6f8d237 ("net: dsa: mt7530: Add MT7621 TRGMII mode support") Tested-by: Bartel Eerdekens <bartel.eerdekens@constell8.be> Signed-off-by: Arınç ÜNAL <arinc.unal@arinc9.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | igc: read before write to SRRCTL registerSong Yoong Siang2023-05-032-5/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | igc_configure_rx_ring() function will be called as part of XDP program setup. If Rx hardware timestamp is enabled prio to XDP program setup, this timestamp enablement will be overwritten when buffer size is written into SRRCTL register. Thus, this commit read the register value before write to SRRCTL register. This commit is tested by using xdp_hw_metadata bpf selftest tool. The tool enables Rx hardware timestamp and then attach XDP program to igc driver. It will display hardware timestamp of UDP packet with port number 9092. Below are detail of test steps and results. Command on DUT: sudo ./xdp_hw_metadata <interface name> Command on Link Partner: echo -n skb | nc -u -q1 <destination IPv4 addr> 9092 Result before this patch: skb hwtstamp is not found! Result after this patch: found skb hwtstamp = 1677800973.642836757 Optionally, read PHC to confirm the values obtained are almost the same: Command: sudo ./testptp -d /dev/ptp0 -g Result: clock time: 1677800973.913598978 or Fri Mar 3 07:49:33 2023 Fixes: fc9df2a0b520 ("igc: Enable RX via AF_XDP zero-copy") Cc: <stable@vger.kernel.org> # 5.14+ Signed-off-by: Song Yoong Siang <yoong.siang.song@intel.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Jesper Dangaard Brouer <brouer@redhat.com> Tested-by: Jesper Dangaard Brouer <brouer@redhat.com> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | pds_core: add AUXILIARY_BUS and NET_DEVLINK to KconfigShannon Nelson2023-05-031-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add selecting of AUXILIARY_BUS and NET_DEVLINK to the pds_core Kconfig. Link: https://lore.kernel.org/netdev/ZE%2FduNH3lBLreNkJ@corigine.com/ Fixes: ddbcb22055d1 ("pds_core: Kconfig and pds_core.rst") Suggested-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: Shannon Nelson <shannon.nelson@amd.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | pds_core: remove CONFIG_DEBUG_FS from makefileShannon Nelson2023-05-031-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | This cruft from previous drafts should have been removed when the code was updated to not use the old style dummy helpers. Fixes: 55435ea7729a ("pds_core: initial framework for pds_core PF driver") Signed-off-by: Shannon Nelson <shannon.nelson@amd.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | ionic: catch failure from devlink_allocShannon Nelson2023-05-031-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a check for NULL on the alloc return. If devlink_alloc() fails and we try to use devlink_priv() on the NULL return, the kernel gets very unhappy and panics. With this fix, the driver load will still fail, but at least it won't panic the kernel. Fixes: df69ba43217d ("ionic: Add basic framework for IONIC Network device driver") Signed-off-by: Shannon Nelson <shannon.nelson@amd.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | r8152: fix the autosuspend doesn't workHayes Wang2023-05-031-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | Set supports_autosuspend = 1 for the rtl8152_cfgselector_driver. Fixes: ec51fbd1b8a2 ("r8152: add USB device driver for config selection") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | ionic: remove noise from ethtool rxnfc error msgShannon Nelson2023-05-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It seems that ethtool is calling into .get_rxnfc more often with ETHTOOL_GRXCLSRLCNT which ionic doesn't know about. We don't need to log a message about it, just return not supported. Fixes: aa3198819bea6 ("ionic: Add RSS support") Signed-off-by: Shannon Nelson <shannon.nelson@amd.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-vf: Detach LF resources on probe cleanupSubbaraya Sundeep2023-05-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a VF device probe fails due to error in MSIX vector allocation then the resources NIX and NPA LFs were not detached. Fix this by detaching the LFs when MSIX vector allocation fails. Fixes: 3184fb5ba96e ("octeontx2-vf: Virtual function driver support") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-pf: Disable packet I/O for graceful exitSubbaraya Sundeep2023-05-031-1/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | At the stage of enabling packet I/O in otx2_open, If mailbox timeout occurs then interface ends up in down state where as hardware packet I/O is enabled. Hence disable packet I/O also before bailing out. Fixes: 1ea0166da050 ("octeontx2-pf: Fix the device state on error") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Skip PFs if not enabledRatheesh Kannoth2023-05-033-11/+46
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Firmware enables PFs and allocate mbox resources for each of the PFs. Currently PF driver configures mbox resources without checking whether PF is enabled or not. This results in crash. This patch fixes this issue by skipping disabled PF's mbox initialization. Fixes: 9bdc47a6e328 ("octeontx2-af: Mbox communication support btw AF and it's VFs") Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Fix issues with NPC field hash extractRatheesh Kannoth2023-05-034-50/+69
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1. Allow field hash configuration for both source and destination IPv6. 2. Configure hardware parser based on hash extract feature enable flag for IPv6. 3. Fix IPv6 endianness issue while updating the source/destination IP address via ntuple rule. Fixes: 56d9f5fd2246 ("octeontx2-af: Use hashed field in MCAM key") Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Update/Fix NPC field hash extract featureRatheesh Kannoth2023-05-033-18/+41
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1. As per previous implementation, mask and control parameter to generate the field hash value was not passed to the caller program. Updated the secret key mbox to share that information as well, as a part of the fix. 2. Earlier implementation did not consider hash reduction of both source and destination IPv6 addresses. Only source IPv6 address was considered. This fix solves that and provides option to hash Fixes: 56d9f5fd2246 ("octeontx2-af: Use hashed field in MCAM key") Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Update correct mask to filter IPv4 fragmentsSuman Ghosh2023-05-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | During the initial design, the IPv4 ip_flag mask was set to 0xff. Which results to filter only fragmets with (fragment_offset == 0). As part of the fix, updated the mask to 0x20 to filter all the fragmented packets irrespective of the fragment_offset value. Fixes: c672e3727989 ("octeontx2-pf: Add support to filter packet based on IP fragment") Signed-off-by: Suman Ghosh <sumang@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Add validation for lmac typeHariprasad Kelam2023-05-031-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Upon physical link change, firmware reports to the kernel about the change along with the details like speed, lmac_type_id, etc. Kernel derives lmac_type based on lmac_type_id received from firmware. In a few scenarios, firmware returns an invalid lmac_type_id, which is resulting in below kernel panic. This patch adds the missing validation of the lmac_type_id field. Internal error: Oops: 96000005 [#1] PREEMPT SMP [ 35.321595] Modules linked in: [ 35.328982] CPU: 0 PID: 31 Comm: kworker/0:1 Not tainted 5.4.210-g2e3169d8e1bc-dirty #17 [ 35.337014] Hardware name: Marvell CN103XX board (DT) [ 35.344297] Workqueue: events work_for_cpu_fn [ 35.352730] pstate: 40400089 (nZcv daIf +PAN -UAO) [ 35.360267] pc : strncpy+0x10/0x30 [ 35.366595] lr : cgx_link_change_handler+0x90/0x180 Fixes: 61071a871ea6 ("octeontx2-af: Forward CGX link notifications to PFs") Signed-off-by: Hariprasad Kelam <hkelam@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-pf: Increase the size of dmac filter flowsRatheesh Kannoth2023-05-031-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | CN10kb supports large number of dmac filter flows to be inserted. Increase the field size to accommodate the same Fixes: b747923afff8 ("octeontx2-af: Exact match support") Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Fix depth of cam and mem table.Ratheesh Kannoth2023-05-031-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In current driver, NPC cam and mem table sizes are read from wrong register offset. This patch fixes the register offset so that correct values are populated on read. Fixes: b747923afff8 ("octeontx2-af: Exact match support") Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Fix start and end bit for scan configRatheesh Kannoth2023-05-031-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the current driver, NPC exact match feature was not getting enabled as configured bit was not read properly. for_each_set_bit_from() need end bit as one bit post position in the bit map to read NPC exact nibble enable bits properly. This patch fixes the same. Fixes: b747923afff8 ("octeontx2-af: Exact match support") Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-af: Secure APR table update with the lockGeetha sowjanya2023-05-031-5/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | APR table contains the lmtst base address of PF/VFs. These entries are updated by the PF/VF during the device probe. The lmtst address is fetched from HW using "TXN_REQ" and "ADDR_RSP_STS" registers. The lock tries to protect these registers from getting overwritten when multiple PFs invokes rvu_get_lmtaddr() simultaneously. For example, if PF1 submit the request and got permitted before it reads the response and PF2 got scheduled submit the request then the response of PF1 is overwritten by the PF2 response. Fixes: 893ae97214c3 ("octeontx2-af: cn10k: Support configurable LMTST regions") Signed-off-by: Geetha sowjanya <gakula@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Sai Krishna <saikrishnag@marvell.com> Reviewed-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: ethernet: mtk_eth_soc: drop generic vlan rx offload, only use DSA untaggingFelix Fietkau2023-05-022-67/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Through testing I found out that hardware vlan rx offload support seems to have some hardware issues. At least when using multiple MACs and when receiving tagged packets on the secondary MAC, the hardware can sometimes start to emit wrong tags on the first MAC as well. In order to avoid such issues, drop the feature configuration and use the offload feature only for DSA hardware untagging on MT7621/MT7622 devices where this feature works properly. Fixes: 08666cbb7dd5 ("net: ethernet: mtk_eth_soc: add support for configuring vlan rx offload") Tested-by: Frank Wunderlich <frank-w@public-files.de> Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Frank Wunderlich <frank-w@public-files.de> Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com> Acked-by: Arınç ÜNAL <arinc.unal@arinc9.com> Link: https://lore.kernel.org/r/20230426172153.8352-1-linux@fw-web.de Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * | net: atlantic: Define aq_pm_ops conditionally on CONFIG_PMTom Rix2023-05-011-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For s390, gcc with W=1 reports drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c:458:32: error: 'aq_pm_ops' defined but not used [-Werror=unused-const-variable=] 458 | static const struct dev_pm_ops aq_pm_ops = { | ^~~~~~~~~ The only use of aq_pm_ops is conditional on CONFIG_PM. The definition of aq_pm_ops and its functions should also be conditional on CONFIG_PM. Signed-off-by: Tom Rix <trix@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | sfc: Fix module EEPROM reporting for QSFP modulesAndy Moreton2023-05-011-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The sfc driver does not report QSFP module EEPROM contents correctly as only the first page is fetched from hardware. Commit 0e1a2a3e6e7d ("ethtool: Add SFF-8436 and SFF-8636 max EEPROM length definitions") added ETH_MODULE_SFF_8436_MAX_LEN for the overall size of the EEPROM info, so use that to report the full EEPROM contents. Fixes: 9b17010da57a ("sfc: Add ethtool -m support for QSFP modules") Signed-off-by: Andy Moreton <andy.moreton@amd.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | r8152: move setting r8153b_rx_agg_chg_indicate()Hayes Wang2023-05-011-6/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Move setting r8153b_rx_agg_chg_indicate() for 2.5G devices. The r8153b_rx_agg_chg_indicate() has to be called after enabling tx/rx. Otherwise, the coalescing settings are useless. Fixes: 195aae321c82 ("r8152: support new chips") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | r8152: fix the poor throughput for 2.5G devicesHayes Wang2023-05-011-0/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the poor throughput for 2.5G devices, when changing the speed from auto mode to force mode. This patch is used to notify the MAC when the mode is changed. Fixes: 195aae321c82 ("r8152: support new chips") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | r8152: fix flow control issue of RTL8156AHayes Wang2023-05-011-20/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The feature of flow control becomes abnormal, if the device sends a pause frame and the tx/rx is disabled before sending a release frame. It causes the lost of packets. Set PLA_RX_FIFO_FULL and PLA_RX_FIFO_EMPTY to zeros before disabling the tx/rx. And, toggle FC_PATCH_TASK before enabling tx/rx to reset the flow control patch and timer. Then, the hardware could clear the state and the flow control becomes normal after enabling tx/rx. Besides, remove inline for fc_pause_on_auto() and fc_pause_off_auto(). Fixes: 195aae321c82 ("r8152: support new chips") Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | net: dsa: mv88e6xxx: add mv88e6321 rsvd2cpuAngelo Dureghello2023-04-281-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add rsvd2cpu capability for mv88e6321 model, to allow proper bpdu processing. Signed-off-by: Angelo Dureghello <angelo.dureghello@timesys.com> Fixes: 51c901a775621 ("net: dsa: mv88e6xxx: distinguish Global 2 Rsvd2CPU") Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | octeontx2-pf: mcs: Do not reset PN while updating secySubbaraya Sundeep2023-04-271-5/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After creating SecYs, SCs and SAs a SecY can be modified to change attributes like validation mode, protect frames mode etc. During this SecY update, packet number is reset to initial user given value by mistake. Hence do not reset PN when updating SecY parameters. Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-pf: mcs: Fix shared counters logicSubbaraya Sundeep2023-04-272-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Macsec stats like InPktsLate and InPktsDelayed share same counter in hardware. If SecY replay_protect is true then counter represents InPktsLate otherwise InPktsDelayed. This mode change was tracked based on protect_frames instead of replay_protect mistakenly. Similarly InPktsUnchecked and InPktsOk share same counter and mode change was tracked based on validate_check instead of validate_disabled. This patch fixes those problems. Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-pf: mcs: Clear stats before freeing resourceSubbaraya Sundeep2023-04-271-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When freeing MCS hardware resources like SecY, SC and SA the corresponding stats needs to be cleared. Otherwise previous stats are shown in newly created macsec interfaces. Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-pf: mcs: Match macsec ethertype along with DMACSubbaraya Sundeep2023-04-271-1/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On CN10KB silicon a single hardware macsec block is present and offloads macsec operations for all the ethernet LMACs. TCAM match with macsec ethertype 0x88e5 alone at RX side is not sufficient to distinguish all the macsec interfaces created on top of netdevs. Hence append the DMAC of the macsec interface too. Otherwise the first created macsec interface only receives all the macsec traffic. Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-pf: mcs: Fix NULL pointer dereferencesSubbaraya Sundeep2023-04-271-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When system is rebooted after creating macsec interface below NULL pointer dereference crashes occurred. This patch fixes those crashes by using correct order of teardown [ 3324.406942] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 [ 3324.415726] Mem abort info: [ 3324.418510] ESR = 0x96000006 [ 3324.421557] EC = 0x25: DABT (current EL), IL = 32 bits [ 3324.426865] SET = 0, FnV = 0 [ 3324.429913] EA = 0, S1PTW = 0 [ 3324.433047] Data abort info: [ 3324.435921] ISV = 0, ISS = 0x00000006 [ 3324.439748] CM = 0, WnR = 0 .... [ 3324.575915] Call trace: [ 3324.578353] cn10k_mdo_del_secy+0x24/0x180 [ 3324.582440] macsec_common_dellink+0xec/0x120 [ 3324.586788] macsec_notify+0x17c/0x1c0 [ 3324.590529] raw_notifier_call_chain+0x50/0x70 [ 3324.594965] call_netdevice_notifiers_info+0x34/0x7c [ 3324.599921] rollback_registered_many+0x354/0x5bc [ 3324.604616] unregister_netdevice_queue+0x88/0x10c [ 3324.609399] unregister_netdev+0x20/0x30 [ 3324.613313] otx2_remove+0x8c/0x310 [ 3324.616794] pci_device_shutdown+0x30/0x70 [ 3324.620882] device_shutdown+0x11c/0x204 [ 966.664930] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 [ 966.673712] Mem abort info: [ 966.676497] ESR = 0x96000006 [ 966.679543] EC = 0x25: DABT (current EL), IL = 32 bits [ 966.684848] SET = 0, FnV = 0 [ 966.687895] EA = 0, S1PTW = 0 [ 966.691028] Data abort info: [ 966.693900] ISV = 0, ISS = 0x00000006 [ 966.697729] CM = 0, WnR = 0 [ 966.833467] Call trace: [ 966.835904] cn10k_mdo_stop+0x20/0xa0 [ 966.839557] macsec_dev_stop+0xe8/0x11c [ 966.843384] __dev_close_many+0xbc/0x140 [ 966.847298] dev_close_many+0x84/0x120 [ 966.851039] rollback_registered_many+0x114/0x5bc [ 966.855735] unregister_netdevice_many.part.0+0x14/0xa0 [ 966.860952] unregister_netdevice_many+0x18/0x24 [ 966.865560] macsec_notify+0x1ac/0x1c0 [ 966.869303] raw_notifier_call_chain+0x50/0x70 [ 966.873738] call_netdevice_notifiers_info+0x34/0x7c [ 966.878694] rollback_registered_many+0x354/0x5bc [ 966.883390] unregister_netdevice_queue+0x88/0x10c [ 966.888173] unregister_netdev+0x20/0x30 [ 966.892090] otx2_remove+0x8c/0x310 [ 966.895571] pci_device_shutdown+0x30/0x70 [ 966.899660] device_shutdown+0x11c/0x204 [ 966.903574] __do_sys_reboot+0x208/0x290 [ 966.907487] __arm64_sys_reboot+0x20/0x30 [ 966.911489] el0_svc_handler+0x80/0x1c0 [ 966.915316] el0_svc+0x8/0x180 [ 966.918362] Code: f9400000 f9400a64 91220014 f94b3403 (f9400060) [ 966.924448] ---[ end trace 341778e799c3d8d7 ]--- Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-af: mcs: Fix MCS block interruptGeetha sowjanya2023-04-274-70/+119
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On CN10KB, MCS IP vector number, BBE and PAB interrupt mask got changed to support more block level interrupts. To address this changes, this patch fixes the bbe and pab interrupt handlers. Fixes: 6c635f78c474 ("octeontx2-af: cn10k: mcs: Handle MCS block interrupts") Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-af: mcs: Config parser to skip 8B headerGeetha sowjanya2023-04-274-0/+41
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When ptp timestamp is enabled in RPM, RPM will append 8B timestamp header for all RX traffic. MCS need to skip these 8 bytes header while parsing the packet header, so that correct tcam key is created for lookup. This patch fixes the mcs parser configuration to skip this 8B header for ptp packets. Fixes: ca7f49ff8846 ("octeontx2-af: cn10k: Introduce driver for macsec block.") Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeontx2-af: mcs: Write TCAM_DATA and TCAM_MASK registers at onceSubbaraya Sundeep2023-04-271-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As per hardware errata on CN10KB, all the four TCAM_DATA and TCAM_MASK registers has to be written at once otherwise write to individual registers will fail. Hence write to all TCAM_DATA registers and then to all TCAM_MASK registers. Fixes: cfc14181d497 ("octeontx2-af: cn10k: mcs: Manage the MCS block hardware resources") Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | octeonxt2-af: mcs: Fix per port bypass configGeetha sowjanya2023-04-272-3/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For each lmac port, MCS has two MCS_TOP_SLAVE_CHANNEL_CONFIGX registers. For CN10KB both register need to be configured for the port level mcs bypass to work. This patch also sets bitmap of flowid/secy entry reserved for default bypass so that these entries can be shown in debugfs. Fixes: bd69476e86fc ("octeontx2-af: cn10k: mcs: Install a default TCAM for normal traffic") Signed-off-by: Geetha sowjanya <gakula@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| * | ixgbe: Fix panic during XDP_TX with > 64 CPUsJohn Hickey2023-04-272-5/+4
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 4fe815850bdc ("ixgbe: let the xdpdrv work with more than 64 cpus") adds support to allow XDP programs to run on systems with more than 64 CPUs by locking the XDP TX rings and indexing them using cpu % 64 (IXGBE_MAX_XDP_QS). Upon trying this out patch on a system with more than 64 cores, the kernel paniced with an array-index-out-of-bounds at the return in ixgbe_determine_xdp_ring in ixgbe.h, which means ixgbe_determine_xdp_q_idx was just returning the cpu instead of cpu % IXGBE_MAX_XDP_QS. An example splat: ========================================================================== UBSAN: array-index-out-of-bounds in /var/lib/dkms/ixgbe/5.18.6+focal-1/build/src/ixgbe.h:1147:26 index 65 is out of range for type 'ixgbe_ring *[64]' ========================================================================== BUG: kernel NULL pointer dereference, address: 0000000000000058 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 [#1] SMP NOPTI CPU: 65 PID: 408 Comm: ksoftirqd/65 Tainted: G IOE 5.15.0-48-generic #54~20.04.1-Ubuntu Hardware name: Dell Inc. PowerEdge R640/0W23H8, BIOS 2.5.4 01/13/2020 RIP: 0010:ixgbe_xmit_xdp_ring+0x1b/0x1c0 [ixgbe] Code: 3b 52 d4 cf e9 42 f2 ff ff 66 0f 1f 44 00 00 0f 1f 44 00 00 55 b9 00 00 00 00 48 89 e5 41 57 41 56 41 55 41 54 53 48 83 ec 08 <44> 0f b7 47 58 0f b7 47 5a 0f b7 57 54 44 0f b7 76 08 66 41 39 c0 RSP: 0018:ffffbc3fcd88fcb0 EFLAGS: 00010282 RAX: ffff92a253260980 RBX: ffffbc3fe68b00a0 RCX: 0000000000000000 RDX: ffff928b5f659000 RSI: ffff928b5f659000 RDI: 0000000000000000 RBP: ffffbc3fcd88fce0 R08: ffff92b9dfc20580 R09: 0000000000000001 R10: 3d3d3d3d3d3d3d3d R11: 3d3d3d3d3d3d3d3d R12: 0000000000000000 R13: ffff928b2f0fa8c0 R14: ffff928b9be20050 R15: 000000000000003c FS: 0000000000000000(0000) GS:ffff92b9dfc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000058 CR3: 000000011dd6a002 CR4: 00000000007706e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: <TASK> ixgbe_poll+0x103e/0x1280 [ixgbe] ? sched_clock_cpu+0x12/0xe0 __napi_poll+0x30/0x160 net_rx_action+0x11c/0x270 __do_softirq+0xda/0x2ee run_ksoftirqd+0x2f/0x50 smpboot_thread_fn+0xb7/0x150 ? sort_range+0x30/0x30 kthread+0x127/0x150 ? set_kthread_struct+0x50/0x50 ret_from_fork+0x1f/0x30 </TASK> I think this is how it happens: Upon loading the first XDP program on a system with more than 64 CPUs, ixgbe_xdp_locking_key is incremented in ixgbe_xdp_setup. However, immediately after this, the rings are reconfigured by ixgbe_setup_tc. ixgbe_setup_tc calls ixgbe_clear_interrupt_scheme which calls ixgbe_free_q_vectors which calls ixgbe_free_q_vector in a loop. ixgbe_free_q_vector decrements ixgbe_xdp_locking_key once per call if it is non-zero. Commenting out the decrement in ixgbe_free_q_vector stopped my system from panicing. I suspect to make the original patch work, I would need to load an XDP program and then replace it in order to get ixgbe_xdp_locking_key back above 0 since ixgbe_setup_tc is only called when transitioning between XDP and non-XDP ring configurations, while ixgbe_xdp_locking_key is incremented every time ixgbe_xdp_setup is called. Also, ixgbe_setup_tc can be called via ethtool --set-channels, so this becomes another path to decrement ixgbe_xdp_locking_key to 0 on systems with more than 64 CPUs. Since ixgbe_xdp_locking_key only protects the XDP_TX path and is tied to the number of CPUs present, there is no reason to disable it upon unloading an XDP program. To avoid confusion, I have moved enabling ixgbe_xdp_locking_key into ixgbe_sw_init, which is part of the probe path. Fixes: 4fe815850bdc ("ixgbe: let the xdpdrv work with more than 64 cpus") Signed-off-by: John Hickey <jjh@daedalian.us> Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://lore.kernel.org/r/20230425170308.2522429-1-anthony.l.nguyen@intel.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* | Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdmaLinus Torvalds2023-04-294-6/+90
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull rdma updates from Jason Gunthorpe: "Usual wide collection of unrelated items in drivers: - Driver bug fixes and treewide cleanups in hfi1, siw, qib, mlx5, rxe, usnic, usnic, bnxt_re, ocrdma, iser: - remove unnecessary NULL checks - kmap obsolescence - pci_enable_pcie_error_reporting() obsolescence - unused variables and macros - trace event related warnings - casting warnings - Code cleanups for irdm and erdma - EFA reporting of 128 byte PCIe TLP support - mlx5 more agressively uses the out of order HW feature - Big rework of how state machines and tasks work in rxe - Fix a syzkaller found crash netdev refcount leak in siw - bnxt_re revises their HW description header - Congestion control for bnxt_re - Use mmu_notifiers more safely in hfi1 - mlx5 gets better support for PCIe relaxed ordering inside VMs" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (81 commits) RDMA/efa: Add rdma write capability to device caps RDMA/mlx5: Use correct device num_ports when modify DC RDMA/irdma: Drop spurious WQ_UNBOUND from alloc_ordered_workqueue() call RDMA/rxe: Fix spinlock recursion deadlock on requester RDMA/mlx5: Fix flow counter query via DEVX RDMA/rxe: Protect QP state with qp->state_lock RDMA/rxe: Move code to check if drained to subroutine RDMA/rxe: Remove qp->req.state RDMA/rxe: Remove qp->comp.state RDMA/rxe: Remove qp->resp.state RDMA/mlx5: Allow relaxed ordering read in VFs and VMs net/mlx5: Update relaxed ordering read HCA capabilities RDMA/mlx5: Check pcie_relaxed_ordering_enabled() in UMR RDMA/mlx5: Remove pcie_relaxed_ordering_enabled() check for RO write RDMA: Add ib_virt_dma_to_page() RDMA/rxe: Fix the error "trying to register non-static key in rxe_cleanup_task" RDMA/irdma: Slightly optimize irdma_form_ah_cm_frame() RDMA/rxe: Fix incorrect TASKLET_STATE_SCHED check in rxe_task.c IB/hfi1: Place struct mmu_rb_handler on cache line start IB/hfi1: Fix bugs with non-PAGE_SIZE-end multi-iovec user SDMA requests ...
| * | RDMA/mlx5: Allow relaxed ordering read in VFs and VMsAvihai Horon2023-04-161-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | According to PCIe spec, Enable Relaxed Ordering value in the VF's PCI config space is wired to 0 and PF relaxed ordering (RO) setting should be applied to the VF. In QEMU (and maybe others), when assigning VFs, the RO bit in PCI config space is not emulated properly and is always set to 0. Therefore, pcie_relaxed_ordering_enabled() always returns 0 for VFs and VMs and thus MKeys can't be created with RO read even if the PF supports it. pcie_relaxed_ordering_enabled() check was added to avoid a syndrome when creating a MKey with relaxed ordering (RO) enabled when the driver's relaxed_ordering_read_pci_enabled HCA capability is out of sync with FW. With the new relaxed_ordering_read capability this can't happen, as it's set regardless of RO value in PCI config space and thus can't change during runtime. Hence, to allow RO read in VFs and VMs, use the new HCA capability relaxed_ordering_read without checking pcie_relaxed_ordering_enabled(). The old capability checks are kept for backward compatibility with older FWs. Allowing RO in VFs and VMs is valuable since it can greatly improve performance on some setups. For example, testing throughput of a VF on an AMD EPYC 7763 and ConnectX-6 Dx setup showed roughly 60% performance improvement. Signed-off-by: Avihai Horon <avihaih@nvidia.com> Reviewed-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Aya Levin <ayal@nvidia.com> Link: https://lore.kernel.org/r/e7048640d66c341a8fa0465e099926e7989184bc.1681131553.git.leon@kernel.org Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
| * | net/mlx5: Update relaxed ordering read HCA capabilitiesAvihai Horon2023-04-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rename existing HCA capability relaxed_ordering_read to relaxed_ordering_read_pci_enabled. This is in accordance with recent PRM change to better describe the capability, as it's set only if both the device supports relaxed ordering (RO) read and RO is enabled in PCI config space. In addition, add new HCA capability relaxed_ordering_read which is set if the device supports RO read, regardless of RO in PCI config space. This will be used in the following patch to allow RO in VFs and VMs. Signed-off-by: Avihai Horon <avihaih@nvidia.com> Reviewed-by: Shay Drory <shayd@nvidia.com> Link: https://lore.kernel.org/r/caa0002fd8135086357dfcc368e2f5cc73b08480.1681131553.git.leon@kernel.org Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
| * | RDMA/mlx5: Remove pcie_relaxed_ordering_enabled() check for RO writeAvihai Horon2023-04-162-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | pcie_relaxed_ordering_enabled() check was added to avoid a syndrome when creating a MKey with relaxed ordering (RO) enabled when the driver's relaxed_ordering_{read,write} HCA capabilities are out of sync with FW. While this can happen with relaxed_ordering_read, it can't happen with relaxed_ordering_write as it's set if the device supports RO write, regardless of RO in PCI config space, and thus can't change during runtime. Therefore, drop the pcie_relaxed_ordering_enabled() check for relaxed_ordering_write while keeping it for relaxed_ordering_read. Doing so will also allow the usage of RO write in VFs and VMs (where RO in PCI config space is not reported/emulated properly). Signed-off-by: Avihai Horon <avihaih@nvidia.com> Reviewed-by: Shay Drory <shayd@nvidia.com> Link: https://lore.kernel.org/r/7e8f55e31572c1702d69cae015a395d3a824a38a.1681131553.git.leon@kernel.org Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
generated by cgit v1.2.3 (git 2.25.1) at 2024-09-15 13:18:07 +0000