- 30 Sep, 2021 16 commits
-
-
Eric Dumazet authored
Using snmp_get_cpu_field_batch() allows for better cpu cache utilization, especially on hosts with large number of cpus. Also remove special handling when mptcp mibs where not yet allocated. I chose to use temporary storage on the stack to keep this patch simple. We might in the future use the storage allocated in netstat_seq_show(). Combined with prior patch (inlining snmp_get_cpu_field) time to fetch and output mptcp counters on a 256 cpu host [1] goes from 75 usec to 16 usec. [1] L1 cache size is 32KB, it is not big enough to hold all dataset. Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Eric Dumazet authored
This trivial function is called ~90,000 times on 256 cpus hosts, when reading /proc/net/netstat. And this number keeps inflating. Inlining it saves many cycles. Signed-off-by:
-
Joshua Roys authored
Add counters for XDP REDIRECT success and failure. This brings the redirect path in line with metrics gathered via the other XDP paths. Signed-off-by:
Joshua Roys <roysjosh@gmail.com> Reviewed-by:
Tariq Toukan <tariqt@nvidia.com> Signed-off-by:
-
Jason Xing authored
Originally, ixgbe driver doesn't allow the mounting of xdpdrv if the server is equipped with more than 64 cpus online. So it turns out that the loading of xdpdrv causes the "NOMEM" failure. Actually, we can adjust the algorithm and then make it work through mapping the current cpu to some xdp ring with the protect of @tx_lock. Here are some numbers before/after applying this patch with xdp-example loaded on the eth0X: As client (tx path): Before After TCP_STREAM send-64 734.14 714.20 TCP_STREAM send-128 1401.91 1395.05 TCP_STREAM send-512 5311.67 5292.84 TCP_STREAM send-1k 9277.40 9356.22 (not stable) TCP_RR send-1 22559.75 21844.22 TCP_RR send-128 23169.54 22725.13 TCP_RR send-512 21670.91 21412.56 As server (rx path): Before After TCP_STREAM send-64 1416.49 1383.12 TCP_STREAM send-128 3141.49 3055.50 TCP_STREAM send-512 9488.73 9487.44 TCP_STREAM send-1k 9491.17 9356.22 (not stable) TCP_RR send-1 23617.74 23601.60 ... Notice: the TCP_RR mode is unstable as the official document explains. I tested many times with different parameters combined through netperf. Though the result is not that accurate, I cannot see much influence on this patch. The static key is places on the hot path, but it actually shouldn't cause a huge regression theoretically. Co-developed-by:Shujin Li <lishujin@kuaishou.com> Signed-off-by:
Jason Xing <xingwanli@kuaishou.com> Tested-by:
Sandeep Penigalapati <sandeep.penigalapati@intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by:
-
David S. Miller authored
Wei Wang says: ==================== net: add new socket option SO_RESERVE_MEM This patch series introduces a new socket option SO_RESERVE_MEM. This socket option provides a mechanism for users to reserve a certain amount of memory for the socket to use. When this option is set, kernel charges the user specified amount of memory to memcg, as well as sk_forward_alloc. This amount of memory is not reclaimable and is available in sk_forward_alloc for this socket. With this socket option set, the networking stack spends less cycles doing forward alloc and reclaim, which should lead to better system performance, with the cost of an amount of pre-allocated and unreclaimable memory, even under memory pressure. With a tcp_stream test with 10 flows running on a simulated 100ms RTT link, I can see the cycles spent in __sk_mem_raise_allocated() dropping by ~0.02%. Not a whole lot, since we already have logic in sk_mem_uncharge() to only reclaim 1MB when sk_forward_alloc has more than 2MB free space. But on a system suffering memory pressure constently, the savings should be more. The first patch is the implementation of this socket option. The following 2 patches change the tcp stack to make use of this reserved memory when under memory pressure. This makes the tcp stack behavior more flexible when under memory pressure, and provides a way for user to control the distribution of the memory among its sockets. With a TCP connection on a simulated 100ms RTT link, the default throughput under memory pressure is ~500Kbps. With SO_RESERVE_MEM set to 100KB, the throughput under memory pressure goes up to ~3.5Mbps. Change since v2: - Added description for new field added in struct sock in patch 1 Change since v1: - Added performance stats in cover letter and rebased ==================== Signed-off-by: -
Wei Wang authored
When user sets SO_RESERVE_MEM socket option, in order to utilize the reserved memory when in memory pressure state, we adjust rcv_ssthresh according to the available reserved memory for the socket, instead of using 4 * advmss always. Signed-off-by:
Wei Wang <weiwan@google.com> Signed-off-by:
-
Wei Wang authored
If user sets SO_RESERVE_MEM socket option, in order to fully utilize the reserved memory in memory pressure state on the tx path, we modify the logic in sk_stream_moderate_sndbuf() to set sk_sndbuf according to available reserved memory, instead of MIN_SOCK_SNDBUF, and adjust it when new data is acked. Signed-off-by:
-
Wei Wang authored
This socket option provides a mechanism for users to reserve a certain amount of memory for the socket to use. When this option is set, kernel charges the user specified amount of memory to memcg, as well as sk_forward_alloc. This amount of memory is not reclaimable and is available in sk_forward_alloc for this socket. With this socket option set, the networking stack spends less cycles doing forward alloc and reclaim, which should lead to better system performance, with the cost of an amount of pre-allocated and unreclaimable memory, even under memory pressure. Note: This socket option is only available when memory cgroup is enabled and we require this reserved memory to be charged to the user's memcg. We hope this could avoid mis-behaving users to abused this feature to reserve a large amount on certain sockets and cause unfairness for others. Signed-off-by:
-
Russell King authored
Add support for the downshift tunable for the Marvell 88x3310 PHY. Downshift is only usable with firmware 0.3.5.0 and later. Signed-off-by:
Russell King <rmk+kernel@armlinux.org.uk> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
Colin Ian King authored
The variable pin is being initialized with a value that is never read, it is being updated later on in only one case of a switch statement. The assignment is redundant and can be removed. Addresses-Coverity: ("Unused value") Signed-off-by:Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Lars-Peter Clausen authored
The macb PTP support currently implements the `gettime64` callback to allow to retrieve the hardware clock time. Update the implementation to provide the `gettimex64` callback instead. The difference between the two is that with `gettime64` a snapshot of the system clock is taken before and after invoking the callback. Whereas `gettimex64` expects the callback itself to take the snapshots. To get the time from the macb Ethernet core multiple register accesses have to be done. Only one of which will happen at the time reported by the function. This leads to a non-symmetric delay and adds a slight offset between the hardware and system clock time when using the `gettime64` method. This offset can be a few 100 nanoseconds. Switching to the `gettimex64` method allows for a more precise correlation of the hardware and system clocks and results in a lower offset between the two. On a Xilinx ZynqMP system `phc2sys` reports a delay of 1120 ns before and 300 ns after the patch. With the latter being mostly symmetric. Signed-off-by:
Lars-Peter Clausen <lars@metafoo.de> Acked-by:
Richard Cochran <richardcochran@gmail.com> Signed-off-by:
-
Boris Sukholitko authored
The following flower filter fails to match non-PPP_IP{V6} packets wrapped in PPP_SES protocol: tc filter add dev eth0 ingress protocol ppp_ses flower \ action simple sdata hi64 The reason is that proto local variable is being set even when FLOW_DISSECT_RET_OUT_BAD status is returned. The fix is to avoid setting proto variable if the PPP protocol is unknown. Signed-off-by:Boris Sukholitko <boris.sukholitko@broadcom.com> Signed-off-by:
-
Linus Walleij authored
We added a state variable to track whether a certain port was VLAN filtering or not, but we can just inquire the DSA core about this. Cc: Vladimir Oltean <olteanv@gmail.com> Cc: Mauri Sandberg <sandberg@mailfence.com> Cc: DENG Qingfang <dqfext@gmail.com> Cc: Alvin Šipraga <alsi@bang-olufsen.dk> Cc: Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
Linus Walleij <linus.walleij@linaro.org> Signed-off-by:
-
Geetha sowjanya authored
Adds XDP_PASS, XDP_TX, XDP_DROP and XDP_REDIRECT support for netdev PF. Signed-off-by:
Geetha sowjanya <gakula@marvell.com> Signed-off-by:
Sunil Goutham <sgoutham@cavium.com> Signed-off-by:
-
Kiran Kumar K authored
In case of ltype NPC_LT_LA_CPT_HDR, LA pointer is pointing to the start of cpt parse header. Since cpt parse header has veriable length padding, this will be a problem for DMAC extraction. Adding KPU profile changes to adjust the LA pointer to start at ether header in case of cpt parse header by - Adding ptr advance in pkind 58 to a fixed value 40 - Adding variable length offset 7 and mask 7 (pad len in CPT_PARSE_HDR). Also added the missing static declaration for npc_set_var_len_offset_pkind function. Signed-off-by:Kiran Kumar K <kirankumark@marvell.com> Signed-off-by:
-
Gustavo A. R. Silva authored
Make use of the struct_size() and flex_array_size() helpers instead of an open-coded version, in order to avoid any potential type mistakes or integer overflows that, in the worse scenario, could lead to heap overflows. Link: https://github.com/KSPP/linux/issues/160Signed-off-by:
Gustavo A. R. Silva <gustavoars@kernel.org> Link: https://lore.kernel.org/r/20210928193107.GA262595@embeddedorSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
- 29 Sep, 2021 24 commits
-
-
Leon Romanovsky authored
The commit mentioned in Fixes line missed a couple of notifications that were registered before devlink_register() and should be delayed too. As such, the too early placed WARN_ON() check spotted it. WARNING: CPU: 1 PID: 6540 at net/core/devlink.c:5158 devlink_nl_region_notify+0x184/0x1e0 net/core/devlink.c:5158 Modules linked in: CPU: 1 PID: 6540 Comm: syz-executor.0 Not tainted 5.15.0-rc2-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:devlink_nl_region_notify+0x184/0x1e0 net/core/devlink.c:5158 Code: 38 41 b8 c0 0c 00 00 31 d2 48 89 ee 4c 89 e7 e8 72 1a 26 00 48 83 c4 08 5b 5d 41 5c 41 5d 41 5e e9 01 bd 41 fa e8 fc bc 41 fa <0f> 0b e9 f7 fe ff ff e8 f0 bc 41 fa 0f 0b eb da 4c 89 e7 e8 c4 18 RSP: 0018:ffffc90002d6f658 EFLAGS: 00010293 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: ffff88801f08d580 RSI: ffffffff87344e94 RDI: 0000000000000003 RBP: ffff88801ee42100 R08: 0000000000000000 R09: 0000000000000000 R10: ffffffff87344d8a R11: 0000000000000000 R12: ffff88801c1dc000 R13: 0000000000000000 R14: 000000000000002c R15: ffff88801c1dc070 FS: 0000555555e8e400(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000055dd7c590310 CR3: 0000000069a09000 CR4: 00000000003506e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: devlink_region_create+0x39f/0x4c0 net/core/devlink.c:10327 nsim_dev_dummy_region_init drivers/net/netdevsim/dev.c:481 [inline] nsim_dev_probe+0x5f6/0x1150 drivers/net/netdevsim/dev.c:1479 call_driver_probe drivers/base/dd.c:517 [inline] really_probe+0x245/0xcc0 drivers/base/dd.c:596 __driver_probe_device+0x338/0x4d0 drivers/base/dd.c:751 driver_probe_device+0x4c/0x1a0 drivers/base/dd.c:781 __device_attach_driver+0x20b/0x2f0 drivers/base/dd.c:898 bus_for_each_drv+0x15f/0x1e0 drivers/base/bus.c:427 __device_attach+0x228/0x4a0 drivers/base/dd.c:969 bus_probe_device+0x1e4/0x290 drivers/base/bus.c:487 device_add+0xc35/0x21b0 drivers/base/core.c:3359 nsim_bus_dev_new drivers/net/netdevsim/bus.c:435 [inline] new_device_store+0x48b/0x770 drivers/net/netdevsim/bus.c:302 bus_attr_store+0x72/0xa0 drivers/base/bus.c:122 sysfs_kf_write+0x110/0x160 fs/sysfs/file.c:139 kernfs_fop_write_iter+0x342/0x500 fs/kernfs/file.c:296 call_write_iter include/linux/fs.h:2163 [inline] new_sync_write+0x429/0x660 fs/read_write.c:507 vfs_write+0x7cf/0xae0 fs/read_write.c:594 ksys_write+0x12d/0x250 fs/read_write.c:647 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7f328409d3ef Code: 89 54 24 18 48 89 74 24 10 89 7c 24 08 e8 99 fd ff ff 48 8b 54 24 18 48 8b 74 24 10 41 89 c0 8b 7c 24 08 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 31 44 89 c7 48 89 44 24 08 e8 cc fd ff ff 48 RSP: 002b:00007ffdc6851140 EFLAGS: 00000293 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f328409d3ef RDX: 0000000000000003 RSI: 00007ffdc6851190 RDI: 0000000000000004 RBP: 0000000000000004 R08: 0000000000000000 R09: 00007ffdc68510e0 R10: 0000000000000000 R11: 0000000000000293 R12: 00007f3284144971 R13: 00007ffdc6851190 R14: 0000000000000000 R15: 00007ffdc6851860 Fixes: cf530217 ("devlink: Notify users when objects are accessible") Reported-by:
Eric Dumazet <eric.dumazet@gmail.com> Tested-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by:
Leon Romanovsky <leonro@nvidia.com> Reviewed-by:
Jacob Keller <jacob.e.keller@intel.com> Link: https://lore.kernel.org/r/2ed1159291f2a589b013914f2b60d8172fc525c1.1632925030.git.leonro@nvidia.comSigned-off-by:
-
Gustavo A. R. Silva authored
Use 2-factor argument form kvcalloc() instead of kvzalloc(). Signed-off-by:
-
Mianhan Liu authored
datagram.c hasn't use any macro or function declared in linux/ip.h. Thus, these files can be removed from datagram.c safely without affecting the compilation of the net/ipv4 module Signed-off-by:
Mianhan Liu <liumh1@shanghaitech.edu.cn> Signed-off-by:
-
Mianhan Liu authored
tag_ksz.c hasn't use any macro or function declared in linux/slab.h. Thus, these files can be removed from tag_ksz.c safely without affecting the compilation of the ./net/dsa module Signed-off-by:
-
Mianhan Liu authored
tag_8021q.c hasn't use any macro or function declared in linux/if_bridge.h. Thus, these files can be removed from tag_8021q.c safely without affecting the compilation of the ./net/dsa module Signed-off-by:
-
Gustavo A. R. Silva authored
Use array_size() helper instead of the open-coded version in copy_to_user(). These sorts of multiplication factors need to be wrapped in array_size(). Link: https://github.com/KSPP/linux/issues/160Signed-off-by:
-
Gustavo A. R. Silva authored
Use array_size() helper instead of the open-coded version in copy_to_user(). These sorts of multiplication factors need to be wrapped in array_size(). Link: https://github.com/KSPP/linux/issues/160Signed-off-by:
-
Gustavo A. R. Silva authored
Use array_size() helper instead of the open-coded version in copy_{from,to}_user(). These sorts of multiplication factors need to be wrapped in array_size(). Link: https://github.com/KSPP/linux/issues/160Signed-off-by: -
David S. Miller authored
Matt Johnston says: ==================== Updates to MCTP core This series adds timeouts for MCTP tags (a limited resource), and a few other improvements to the MCTP core. ==================== Signed-off-by: -
Matt Johnston authored
Should not occur but is a sanity check. May help tracking down Trinity reported issue https://lore.kernel.org/lkml/20210913030701.GA5926@xsang-OptiPlex-9020/Signed-off-by:
Matt Johnston <matt@codeconstruct.com.au> Signed-off-by:
-
Matt Johnston authored
A route's RTAX_MTU can be set in nested RTAX_METRICS Signed-off-by:
-
Jeremy Kerr authored
Describe common flows and refcounting behaviour. Signed-off-by:
Jeremy Kerr <jk@codeconstruct.com.au> Signed-off-by:
-
Jeremy Kerr authored
In a future change, we'll want to provide a registration call for mctp-specific devices. This requires us to have the networks established before device driver inits, so run the core init as a subsys_initcall. Signed-off-by:
-
Jeremy Kerr authored
The tag allocation, release and bind events are somewhat opaque outside the kernel; this change adds a few tracepoints to assist in instrumentation and debugging. Signed-off-by:
-
Jeremy Kerr authored
Currently, a MCTP (local-eid,remote-eid,tag) tuple is allocated to a socket on send, and only expires when the socket is closed. This change introduces a tag timeout, freeing the tuple after a fixed expiry - currently six seconds. This is greater than (but close to) the max response timeout in upper-layer bindings. Signed-off-by:
-
Jeremy Kerr authored
Currently, we tie the struct mctp_dev lifetime to the underlying struct net_device, and hold/put that device as a proxy for a separate mctp_dev refcount. This works because we're not holding any references to the mctp_dev that are different from the netdev lifetime. In a future change we'll break that assumption though, as we'll need to hold mctp_dev references in a workqueue, which might live past the netdev unregister notification. In order to support that, this change introduces a refcount on the mctp_dev, currently taken by the net_device->mctp_ptr reference, and released on netdev unregister events. We can then use this for future references that might outlast the net device. Signed-off-by:
-
Jeremy Kerr authored
We will want to invalidate sk_keys in a future change, which will require a boolean flag to mark invalidated items in the socket & net namespace lists. We'll also need to take a reference to keys, held over non-atomic contexts, so we need a refcount on keys also. This change adds a validity flag (currently always true) and refcount to struct mctp_sk_key. With a refcount on the keys, using RCU no longer makes much sense; we have exact indications on the lifetime of keys. So, we also change the RCU list traversal to a locked implementation. Signed-off-by:
-
Jeremy Kerr authored
We may need to receive packets addressed to the null EID (==0), but addressed to us at the physical layer. This change adds a lookup for local routes when we see a packet addressed to EID 0, and a local phys address. Signed-off-by:
-
Matt Johnston authored
Allowing TUN is useful for testing, to route packets to userspace or to tunnel between machines. Signed-off-by:
-
Horatiu Vultur authored
The LAN8804 PHY has same features as that of LAN8814 PHY except that it doesn't support 1588, SyncE or Q-USGMII. This PHY is found inside the LAN966X switches. Reviewed-by:
Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/nexDavid S. Miller authored
t-queue Tony Nguyen says: ==================== 100GbE Intel Wired LAN Driver Updates 2021-09-28 This series contains updates to ice driver only. Dave adds support for QoS DSCP allowing for DSCP to TC mapping via APP TLVs. Ani adds enforcement of DSCP to only supported devices with the introduction of a feature bitmap and corrects messaging of unsupported modules based on link mode. Jake refactors devlink info functions to be void as the functions no longer return errors. Jeff fixes a macro name to properly reflect the value. Len Baker converts a kzalloc allocation to, the preferred, kcalloc. ==================== Signed-off-by: -
David S. Miller authored
Subbaraya Sundeep <sbhatta@marvell.com> ==================== octeontx2: Add PTP support for VFs PTP is a shared hardware block which can prepend RX timestamps to packets before directing packets to PFs or VFs and can notify the TX timestamps to PFs or VFs via TX completion queue descriptors. Hence adding PTP support for VFs is exactly similar to PFs with minimal changes. This patchset adds that PTP support for VFs. Patch 1 - When an interface is set in promisc/multicast the same setting is not retained when changing mtu or channels. This is due to toggling of the interface by driver but not calling set_rx_mode in the down-up sequence. Since setting an interface to multicast properly is required for ptp this is addressed in this patch. Patch 2 - Changes in VF driver for registering timestamping ethtool ops and ndo_ioctl. ==================== Signed-off-by: -
Naveen Mamindlapalli authored
This patch adds PTP PHC support to NIX VF interfaces. This enables a VF to run PTP master/slave instance. PTP block being a shared hardware resource it is recommended to avoid running multiple PTP instances in the system which will impact the PTP clock accuracy. Signed-off-by:
Naveen Mamindlapalli <naveenm@marvell.com> Signed-off-by:
Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by:
Sunil Kovvuri Goutham <sgoutham@marvell.com> Signed-off-by:
-
Rakesh Babu authored
Whenever the interface is brought up/down then set_rx_mode function is called by the stack which enables promisc/allmulti MCAM entries. But there are cases when driver brings interface down and then up such as while changing number of channels. In these cases promisc/allmulti MCAM entries are left disabled as set_rx_mode callback is not called. This patch enables these MCAM entries in all such cases. Signed-off-by:
Rakesh Babu <rsaladi2@marvell.com> Signed-off-by:
-
