- 23 Apr, 2021 20 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-nextDavid S. Miller authored
Steffen Klassert says: ==================== pull request (net-next): ipsec-next 2021-04-23 1) The SPI flow key in struct flowi has no consumers, so remove it. From Florian Westphal. 2) Remove stray synchronize_rcu from xfrm_init. From Florian Westphal. 3) Use the new exit_pre hook to reset the netlink socket on net namespace destruction. From Florian Westphal. 4) Remove an unnecessary get_cpu() in ipcomp, that code is always called with BHs off. From Sabrina Dubroca. Please pull or let me know if there are problems. ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Ilya Lipnitskiy says: ==================== mtk_eth_soc: fixes and performance improvements Most of these changes come from OpenWrt where they have been present and tested for months. First three patches are bug fixes. The rest are performance improvements. The last patch is a cleanup to use the iopoll.h macro for busy-waiting instead of a custom loop. v2: - Reverse christmas tree in "use iopoll.h macro for DMA init" - Use cond_resched() instead of iopoll.h macro in "reduce MDIO bus access latency" - Use napi_complete_done and rework NAPI callbacks in a new patch ==================== Signed-off-by: -
Ilya Lipnitskiy authored
Replace a tight busy-wait loop without a pause with a standard readx_poll_timeout_atomic routine with a 5 us poll period. Tested by booting a MT7621 device to ensure the driver initializes properly. Signed-off-by:
Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
Felix Fietkau authored
This improves GRO performance Signed-off-by:
Felix Fietkau <nbd@nbd.name> [Ilya: Use MTK_RXD4_FOE_ENTRY instead of GENMASK(13, 0)] Signed-off-by:
-
Ilya Lipnitskiy authored
Use napi_complete_done to communicate total TX and RX work done to NAPI. Count total RX work up instead of remaining work down for clarity. Remove unneeded local variables for clarity. Use do {} while instead of goto for clarity. Suggested-by:Jakub Kicinski <kuba@kernel.org> Signed-off-by:
-
Felix Fietkau authored
Avoid rearming interrupt if napi_complete returns false Signed-off-by:
-
Felix Fietkau authored
Uncached memory access is expensive, and there is no need to access all descriptor words if we can't process them anyway Signed-off-by:
-
Felix Fietkau authored
The value is only updated by the CPU, so it is cheaper to access from the ring data structure than from a hardware register. Signed-off-by:
-
Felix Fietkau authored
Reduces the number of interrupts under load Signed-off-by:
-
Felix Fietkau authored
256 descriptors is not enough for multi-gigabit traffic under load on MT7622. Bump it to 512 to improve performance. Signed-off-by:
-
Felix Fietkau authored
Improves tx performance Signed-off-by:
-
Felix Fietkau authored
When running short on descriptors, only stop the queue for the netdev that tx was attempted for. By the time something tries to send on the other netdev, the ring might have some more room already. Signed-off-by:
-
Felix Fietkau authored
usleep_range often ends up sleeping much longer than the 10-20us provided as a range here. This causes significant latency in mdio bus acceses, which easily adds multiple seconds to the boot time on MT7621 when polling DSA slave ports. Use cond_resched instead of usleep_range, since the MDIO access does not take much time Signed-off-by:
-
Felix Fietkau authored
Should improve performance, since it can use bulk free Signed-off-by:
-
Ilya Lipnitskiy authored
In case build_skb fails, call skb_free_frag on the correct pointer. Also update the DMA structures with the new mapping before exiting, because the mapping was successful Suggested-by:
-
Felix Fietkau authored
Since build_skb accesses the data area (for initializing shinfo), dma unmap needs to happen before that call Signed-off-by:
-
Felix Fietkau authored
The VLAN ID in the rx descriptor is only valid if the RX_DMA_VTAG bit is set. Fixes frames wrongly marked with VLAN tags. Signed-off-by:
-
Dexuan Cui authored
mana_gd_poll_cq() may return -1 if an overflow error is detected (this should never happen unless there is a bug in the driver or the hardware). Fix the type of the variable "comp_read" by using int rather than u32. Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Fixes: ca9c54d2 ("net: mana: Add a driver for Microsoft Azure Network Adapter (MANA)") Signed-off-by:
Dexuan Cui <decui@microsoft.com> Signed-off-by:
-
Xuan Zhuo authored
When "headroom" > 0, the actual allocated memory space is the entire page, so the address of the page should be used when passing it to build_skb(). BUG: KASAN: use-after-free in skb_gro_receive (net/core/skbuff.c:4260) Write of size 16 at addr ffff88811619fffc by task kworker/u9:0/534 CPU: 2 PID: 534 Comm: kworker/u9:0 Not tainted 5.12.0-rc7-custom-16372-gb150be05b806 #3382 Hardware name: QEMU MSN2700, BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Workqueue: xprtiod xs_stream_data_receive_workfn [sunrpc] Call Trace: <IRQ> dump_stack (lib/dump_stack.c:122) print_address_description.constprop.0 (mm/kasan/report.c:233) kasan_report.cold (mm/kasan/report.c:400 mm/kasan/report.c:416) skb_gro_receive (net/core/skbuff.c:4260) tcp_gro_receive (net/ipv4/tcp_offload.c:266 (discriminator 1)) tcp4_gro_receive (net/ipv4/tcp_offload.c:316) inet_gro_receive (net/ipv4/af_inet.c:1545 (discriminator 2)) dev_gro_receive (net/core/dev.c:6075) napi_gro_receive (net/core/dev.c:6168 net/core/dev.c:6198) receive_buf (drivers/net/virtio_net.c:1151) virtio_net virtnet_poll (drivers/net/virtio_net.c:1415 drivers/net/virtio_net.c:1519) virtio_net __napi_poll (net/core/dev.c:6964) net_rx_action (net/core/dev.c:7033 net/core/dev.c:7118) __do_softirq (./arch/x86/include/asm/jump_label.h:25 ./include/linux/jump_label.h:200 ./include/trace/events/irq.h:142 kernel/softirq.c:346) irq_exit_rcu (kernel/softirq.c:221 kernel/softirq.c:422 kernel/softirq.c:434) common_interrupt (arch/x86/kernel/irq.c:240 (discriminator 14)) </IRQ> Fixes: fb32856b ("virtio-net: page_to_skb() use build_skb when there's sufficient tailroom") Signed-off-by:
Xuan Zhuo <xuanzhuo@linux.alibaba.com> Reported-by:
Ido Schimmel <idosch@nvidia.com> Tested-by:
Jason Wang <jasowang@redhat.com> Signed-off-by:
-
Tonghao Zhang authored
tw_prot_cleanup will check the twsk_prot. Fixes: 0f5907af ("net: Fix potential memory leak in proto_register()") Cc: Miaohe Lin <linmiaohe@huawei.com> Signed-off-by:
Tonghao Zhang <xiangxia.m.yue@gmail.com> Signed-off-by:
-
- 22 Apr, 2021 20 commits
-
-
David S. Miller authored
Mohammad Athari Bin Ismail says: ==================== Enable DWMAC HW descriptor prefetch This patch series to add setting for HW descriptor prefetch for DWMAC version 5.20 onwards. For Intel platform, enable the capability by default. ==================== Signed-off-by: -
Mohammad Athari Bin Ismail authored
Enable HW descriptor prefetch by default by setting plat->dma_cfg->dche = true in intel_mgbe_common_data(). Need to be noted that this capability only be supported in DWMAC core version 5.20 onwards. In stmmac, there is a checking to check the core version. If the core version is below 5.20, this capability wouldn`t be configured. Below is the iperf result comparison between HW descriptor prefetch disabled(DCHE=0b) and enabled(DCHE=1b). Tested on Intel Elkhartlake platform with DWMAC Core 5.20. Observed line rate performance improvement with HW descriptor prefetch enabled. DCHE = 0b [ 5] local 169.254.1.162 port 42123 connected to 169.254.244.142 port 5201 [ ID] Interval Transfer Bitrate Total Datagrams [ 5] 0.00-1.00 sec 96.7 MBytes 811 Mbits/sec 70050 [ 5] 1.00-2.00 sec 96.5 MBytes 809 Mbits/sec 69850 [ 5] 2.00-3.00 sec 96.3 MBytes 808 Mbits/sec 69720 [ 5] 3.00-4.00 sec 95.9 MBytes 804 Mbits/sec 69450 [ 5] 4.00-5.00 sec 96.0 MBytes 806 Mbits/sec 69530 [ 5] 5.00-6.00 sec 96.8 MBytes 812 Mbits/sec 70080 [ 5] 6.00-7.00 sec 96.9 MBytes 813 Mbits/sec 70140 [ 5] 7.00-8.00 sec 96.8 MBytes 812 Mbits/sec 70080 [ 5] 8.00-9.00 sec 97.0 MBytes 814 Mbits/sec 70230 [ 5] 9.00-10.00 sec 96.9 MBytes 813 Mbits/sec 70170 - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate Jitter Lost/Total Datagrams [ 5] 0.00-10.00 sec 966 MBytes 810 Mbits/sec 0.000 ms 0/699300 (0%) sender [ 5] 0.00-10.00 sec 966 MBytes 810 Mbits/sec 0.011 ms 0/699265 (0%) receiver DCHE = 1b [ 5] local 169.254.1.162 port 49740 connected to 169.254.244.142 port 5201 [ ID] Interval Transfer Bitrate Total Datagrams [ 5] 0.00-1.00 sec 97.9 MBytes 821 Mbits/sec 70880 [ 5] 1.00-2.00 sec 98.1 MBytes 823 Mbits/sec 71060 [ 5] 2.00-3.00 sec 98.2 MBytes 824 Mbits/sec 71140 [ 5] 3.00-4.00 sec 98.2 MBytes 824 Mbits/sec 71090 [ 5] 4.00-5.00 sec 98.1 MBytes 823 Mbits/sec 71050 [ 5] 5.00-6.00 sec 98.1 MBytes 823 Mbits/sec 71040 [ 5] 6.00-7.00 sec 98.1 MBytes 823 Mbits/sec 71050 [ 5] 7.00-8.00 sec 98.2 MBytes 824 Mbits/sec 71140 [ 5] 8.00-9.00 sec 98.2 MBytes 824 Mbits/sec 71120 [ 5] 9.00-10.00 sec 98.3 MBytes 824 Mbits/sec 71150 - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate Jitter Lost/Total Datagrams [ 5] 0.00-10.00 sec 981 MBytes 823 Mbits/sec 0.000 ms 0/710720 (0%) sender [ 5] 0.00-10.00 sec 981 MBytes 823 Mbits/sec 0.041 ms 0/710650 (0%) receiver Signed-off-by:
Mohammad Athari Bin Ismail <mohammad.athari.ismail@intel.com> Signed-off-by:
-
Mohammad Athari Bin Ismail authored
DWMAC Core 5.20 onwards supports HW descriptor prefetching. Additionally, it also depends on platform specific RTL configuration. This capability could be enabled by setting DMA_Mode bit-19 (DCHE). So, to enable this cability, platform must set plat->dma_cfg->dche = true and the DWMAC core version must be 5.20 onwards. Else, this capability wouldn`t be configured Signed-off-by:
-
Hans Westgaard Ry authored
Handling comm_channel_event in mlx4_master_comm_channel uses a double loop to determine which slaves have requested work. The search is always started at lowest slave. This leads to unfairness; lower VFs tends to be prioritized over higher VFs. The patch uses find_next_bit to determine which slaves to handle. Fairness is implemented by always starting at the next to the last start. An MPI program has been used to measure improvements. It runs 500 ibv_reg_mr, synchronizes with all other instances and then runs 500 ibv_dereg_mr. The results running 500 processes, time reported is for running 500 calls: ibv_reg_mr: Mod. Org. mlx4_1 403.356ms 424.674ms mlx4_2 403.355ms 424.674ms mlx4_3 403.354ms 424.674ms mlx4_4 403.355ms 424.674ms mlx4_5 403.357ms 424.677ms mlx4_6 403.354ms 424.676ms mlx4_7 403.357ms 424.675ms mlx4_8 403.355ms 424.675ms ibv_dereg_mr: Mod. Org. mlx4_1 116.408ms 142.818ms mlx4_2 116.434ms 142.793ms mlx4_3 116.488ms 143.247ms mlx4_4 116.679ms 143.230ms mlx4_5 112.017ms 107.204ms mlx4_6 112.032ms 107.516ms mlx4_7 112.083ms 184.195ms mlx4_8 115.089ms 190.618ms Suggested-by:Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by:
Hans Westgaard Ry <hans.westgaard.ry@oracle.com> Reviewed-by:
Tariq Toukan <tariqt@nvidia.com> Signed-off-by:
-
Hayes Wang authored
When the timeout occurs, we still have to run the following process for releasing patch request. Otherwise, the PHY would keep no link. Therefore, use break to stop the loop of loading firmware and release the patch request rather than return the function directly. Fixes: 4a51b0e8 ("r8152: support PHY firmware for RTL8156 series") Signed-off-by:
Hayes Wang <hayeswang@realtek.com> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queueDavid S. Miller authored
Tony Nguyen says: ==================== 100GbE Intel Wired LAN Driver Updates 2021-04-22 This series contains updates to virtchnl header file, ice, and iavf drivers. Vignesh adds support to warn about potentially malicious VFs; those that are overflowing the mailbox for the ice driver. Michal adds support for an allowlist/denylist of VF commands based on supported capabilities for the ice driver. Brett adds support for iavf UDP segmentation offload by adding the capability bit to virtchnl, advertising support in the ice driver, and enabling it in the iavf driver. He also adds a helper function for getting the VF VSI for ice. Colin Ian King removes an unneeded pointer assignment. Qi enables support in the ice driver to support virtchnl requests from the iavf to configure its own RSS input set. This includes adding new capability bits, structures, and commands to virtchnl header file. Haiyue enables configuring RSS flow hash via ethtool to support TCP, UDP and SCTP protocols in iavf. ==================== Signed-off-by: -
Arnd Bergmann authored
There are a few warnings about empty debug macros in this driver: drivers/net/ethernet/neterion/vxge/vxge-main.c: In function 'vxge_probe': drivers/net/ethernet/neterion/vxge/vxge-main.c:4480:76: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body] 4480 | "Failed in enabling SRIOV mode: %d\n", ret); Change them to proper 'do { } while (0)' expressions to make the code a little more robust and avoid the warnings. Signed-off-by:Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Loic Poulain authored
Ensure that the poll system call returns proper error flags when port is removed (nullified port ops), allowing user side to properly fail, without further read or write. Fixes: 9a44c1cc ("net: Add a WWAN subsystem") Signed-off-by:
Loic Poulain <loic.poulain@linaro.org> Reviewed-by:
Leon Romanovsky <leonro@nvidia.com> Signed-off-by:
-
Ido Schimmel authored
When the sampling truncation length is invalid (zero), pass the length of the packet. Without the fix, no payload is reported to user space when the truncation length is zero. Fixes: a8700c3d ("netdevsim: Add dummy psample implementation") Signed-off-by:
-
Arnd Bergmann authored
A link time bug that I had fixed before has come back now that another sub-module was added to the enetc driver: ERROR: modpost: "enetc_ierb_register_pf" [drivers/net/ethernet/freescale/enetc/fsl-enetc.ko] undefined! The problem is that the enetc Makefile is not actually used for the ierb module if that is the only built-in driver in there and everything else is a loadable module. Fix it by always entering the directory this time, regardless of which symbols are configured. This should reliably fix the problem and prevent it from coming back another time. Fixes: 112463dd ("net: dsa: felix: fix link error") Fixes: e7d48e5f ("net: enetc: add a mini driver for the Integrated Endpoint Register Block") Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by:
-
Arnd Bergmann authored
The MANA driver causes a build failure in some configurations when it selects an unavailable symbol: WARNING: unmet direct dependencies detected for PCI_HYPERV Depends on [n]: PCI [=y] && X86_64 [=y] && HYPERV [=n] && PCI_MSI [=y] && PCI_MSI_IRQ_DOMAIN [=y] && SYSFS [=y] Selected by [y]: - MICROSOFT_MANA [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_MICROSOFT [=y] && PCI_MSI [=y] && X86_64 [=y] drivers/pci/controller/pci-hyperv.c: In function 'hv_irq_unmask': drivers/pci/controller/pci-hyperv.c:1217:9: error: implicit declaration of function 'hv_set_msi_entry_from_desc' [-Werror=implicit-function-declaration] 1217 | hv_set_msi_entry_from_desc(¶ms->int_entry.msi_entry, msi_desc); | ^~~~~~~~~~~~~~~~~~~~~~~~~~ A PCI driver should never depend on a particular host bridge implementation in the first place, but if we have this dependency it's better to express it as a 'depends on' rather than 'select'. Signed-off-by: -
Haiyue Wang authored
Provide the ability to enable SCTP RSS hashing by ethtool. It gives users option of generating RSS hash based on the SCTP source and destination ports numbers, IPv4 or IPv6 source and destination addresses. Signed-off-by:
Haiyue Wang <haiyue.wang@intel.com> Tested-by:
Konrad Jankowski <konrad0.jankowski@intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com>
-
Haiyue Wang authored
Provides the ability to enable UDP RSS hashing by ethtool. It gives users option of generating RSS hash based on the UDP source and destination ports numbers, IPv4 or IPv6 source and destination addresses. Signed-off-by:
-
Haiyue Wang authored
Provides the ability to enable TCP RSS hashing by ethtool. It gives users option of generating RSS hash based on the TCP source and destination ports numbers, IPv4 or IPv6 source and destination addresses. Signed-off-by:
-
Haiyue Wang authored
Add the virtchnl message interface to VF, so that VF can request RSS input set(s) based on PF's capability. This framework allows ethtool RSS config support on the VF driver. Signed-off-by:
-
Qi Zhang authored
Add the handler for virtchnl message VIRTCHNL_OP_DEL_RSS_CFG to remove an existing RSS configuration with matching hashed fields. Signed-off-by:
Vignesh Sridhar <vignesh.sridhar@intel.com> Co-developed-by:
Jia Guo <jia.guo@intel.com> Signed-off-by:
Qi Zhang <qi.z.zhang@intel.com> Signed-off-by:
Bo Chen <BoX.C.Chen@intel.com> Signed-off-by:
-
Qi Zhang authored
Currently, RSS hash input is not available to AVF by ethtool, it is set by the PF directly. Add the RSS configure support for AVF through new virtchnl message, and define the capability flag VIRTCHNL_VF_OFFLOAD_ADV_RSS_PF to query this new RSS offload support. Signed-off-by:
-
Brett Creeley authored
Currently, the driver gets the VF's VSI by using a long string of dereferences (i.e. vf->pf->vsi[vf->lan_vsi_idx]). If the method to get the VF's VSI were to change the driver would have to change it in every location. Fix this by adding the helper ice_get_vf_vsi(). Signed-off-by:
Brett Creeley <brett.creeley@intel.com> Tested-by:
-
Colin Ian King authored
Pointer vsi is being re-assigned a value that is never read, the assignment is redundant and can be removed. Addresses-Coverity: ("Unused value") Signed-off-by:Colin Ian King <colin.king@canonical.com> Tested-by:
Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by:
-
Brett Creeley authored
Add code to support UDP segmentation offload (USO) for hardware that supports it. Suggested-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by:
-
