- 05 Mar, 2024 40 commits
-
-
Jakub Kicinski authored
Ido Schimmel says: ==================== selftests: forwarding: Various improvements This patchset speeds up the multipath tests (patches #1-#2) and makes other tests more stable (patches #3-#6) so that they will not randomly fail in the netdev CI. On my system, after applying the first two patches, the run time of gre_multipath_nh_res.sh is reduced by over 90%. ==================== Link: https://lore.kernel.org/r/20240304095612.462900-1-idosch@nvidia.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Ido Schimmel authored
These tests generate various IPv6 flows, encapsulate them in GRE packets and check that the encapsulated packets are distributed between the available nexthops according to the configured weights. Unlike the corresponding IPv4 tests, these tests sometimes fail in the netdev CI because of large discrepancies between the expected and measured ratios [1]. This can be explained by the fact that the IPv4 tests generate about 3,600 different flows whereas the IPv6 tests only generate about 784 different flows (potentially by mistake). Fix by aligning the IPv6 tests to the IPv4 ones and increase the number of generated flows. [1] [...] # TEST: ping [ OK ] # INFO: Running IPv6 over GRE over IPv4 multipath tests # TEST: ECMP [FAIL] # Too large discrepancy between expected and measured ratios # INFO: Expected ratio 1.00 Measured ratio 1.18 [...] Signed-off-by:
Ido Schimmel <idosch@nvidia.com> Link: https://lore.kernel.org/r/20240304095612.462900-7-idosch@nvidia.comSigned-off-by:
-
Ido Schimmel authored
These tests sometimes fail on the netdev CI because the expected number of packets is larger than expected [1]. Make the tests more robust by specifically matching on VXLAN encapsulated packets and allowing up to five stray packets instead of just two. [1] [...] # TEST: VXLAN: ECN encap: 0x00->0x00 [FAIL] # v1: Expected to capture 10 packets, got 13. # TEST: VXLAN: ECN encap: 0x01->0x01 [ OK ] # TEST: VXLAN: ECN encap: 0x02->0x02 [ OK ] # TEST: VXLAN: ECN encap: 0x03->0x02 [ OK ] [...] Signed-off-by:
-
Ido Schimmel authored
The ageing time used by the test is too short for debug kernels and results in entries being aged out prematurely [1]. Fix by increasing the ageing time. [1] # ./vxlan_bridge_1q.sh [...] INFO: learning vlan 10 TEST: VXLAN: flood before learning [ OK ] TEST: VXLAN: show learned FDB entry [ OK ] TEST: VXLAN: learned FDB entry [FAIL] swp4: Expected to capture 0 packets, got 10. RTNETLINK answers: No such file or directory TEST: VXLAN: deletion of learned FDB entry [ OK ] TEST: VXLAN: Ageing of learned FDB entry [FAIL] swp4: Expected to capture 0 packets, got 10. TEST: VXLAN: learning toggling on bridge port [ OK ] [...] Signed-off-by: -
Ido Schimmel authored
The test configures a policer with a rate of 80Mbps and expects to measure a rate close to it. This is a too high rate for debug kernels, causing the test to fail [1]. Fix by reducing the rate to 10Mbps. [1] # ./tc_police.sh TEST: police on rx [FAIL] Expected rate 76.2Mbps, got 29.6Mbps, which is -61% off. Required accuracy is +-10%. TEST: police on tx [FAIL] Expected rate 76.2Mbps, got 30.4Mbps, which is -60% off. Required accuracy is +-10%. Signed-off-by: -
Ido Schimmel authored
The various multipath tests use mausezahn to generate different flows and check how they are distributed between the available nexthops. The tool is currently invoked with an hard coded transmission delay of 1 ms. This is unnecessary when the tests are run with veth pairs and needlessly prolongs the tests. Parametrize this delay and default it to 0 us. It can be overridden using the forwarding.config file. On my system, this reduces the run time of router_multipath.sh by 93%. Signed-off-by:
-
Ido Schimmel authored
The multipath tests currently test both the L3 and L4 multipath hash policies for IPv6, but only the L4 policy for IPv4. The reason is mostly historic: When the initial multipath test was added (router_multipath.sh) the IPv6 L4 policy did not exist and was later added to the test. The other multipath tests copied this pattern although there is little value in testing both policies. Align the IPv4 and IPv6 tests and only test the L4 policy. On my system, this reduces the run time of router_multipath.sh by 89% because of the repeated ping6 invocations to randomize the flow label. Signed-off-by:
-
Eric Dumazet authored
Many syzbot reports show extreme rtnl pressure, and many of them hint that smc acquires rtnl in netns creation for no good reason [1] This patch returns early from smc_pnet_net_init() if there is no netdevice yet. I am not even sure why smc_pnet_create_pnetids_list() even exists, because smc_pnet_netdev_event() is also calling smc_pnet_add_base_pnetid() when handling NETDEV_UP event. [1] extract of typical syzbot reports 2 locks held by syz-executor.3/12252: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.4/12253: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.1/12257: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.2/12261: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.0/12265: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.3/12268: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.4/12271: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.1/12274: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 2 locks held by syz-executor.2/12280: #0: ffffffff8f369610 (pernet_ops_rwsem){++++}-{3:3}, at: copy_net_ns+0x4c7/0x7b0 net/core/net_namespace.c:491 #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_create_pnetids_list net/smc/smc_pnet.c:809 [inline] #1: ffffffff8f375b88 (rtnl_mutex){+.+.}-{3:3}, at: smc_pnet_net_init+0x10a/0x1e0 net/smc/smc_pnet.c:878 Signed-off-by:Eric Dumazet <edumazet@google.com> Cc: Wenjia Zhang <wenjia@linux.ibm.com> Cc: Jan Karcher <jaka@linux.ibm.com> Cc: "D. Wythe" <alibuda@linux.alibaba.com> Cc: Tony Lu <tonylu@linux.alibaba.com> Cc: Wen Gu <guwen@linux.alibaba.com> Reviewed-by:
Wenjia Zhang <wenjia@linux.ibm.com> Link: https://lore.kernel.org/r/20240302100744.3868021-1-edumazet@google.comSigned-off-by:
Paolo Abeni <pabeni@redhat.com>
-
Paolo Abeni authored
Merge tag 'linux-can-next-for-6.9-20240304' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next Marc Kleine-Budde says: ==================== pull-request: can-next 2024-03-04 this is a pull request of 4 patches for net-next/master. The 1st patch is by Jimmy Assarsson and adds support for the Leaf v3 to the kvaser_usb driver. Martin Jocić's patch targets the kvaser_pciefd driver and adds support for the Kvaser PCIe 8xCAN device. Followed by a patch by me that adds a missing a cpu_to_le32() to the gs_usb driver, the change is not critical as the assigned value is 0. The last patch is also by me and replaces a literal 256 with a proper define. linux-can-next-for-6.9-20240304 * tag 'linux-can-next-for-6.9-20240304' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next: can: mcp251xfd: __mcp251xfd_get_berr_counter(): use CAN_BUS_OFF_THRESHOLD instead of open coding it can: gs_usb: gs_cmd_reset(): use cpu_to_le32() to assign mode can: kvaser_pciefd: Add support for Kvaser PCIe 8xCAN can: kvaser_usb: Add support for Leaf v3 ==================== Link: https://lore.kernel.org/r/20240304092051.3631481-1-mkl@pengutronix.deSigned-off-by:
-
Abhishek Chauhan authored
Bridge driver today has no support to forward the userspace timestamp packets and ends up resetting the timestamp. ETF qdisc checks the packet coming from userspace and encounters to be 0 thereby dropping time sensitive packets. These changes will allow userspace timestamps packets to be forwarded from the bridge to NIC drivers. Setting the same bit (mono_delivery_time) to avoid dropping of userspace tstamp packets in the forwarding path. Existing functionality of mono_delivery_time remains unaltered here, instead just extended with userspace tstamp support for bridge forwarding path. Signed-off-by:
Abhishek Chauhan <quic_abchauha@quicinc.com> Reviewed-by:
Willem de Bruijn <willemb@google.com> Link: https://lore.kernel.org/r/20240301201348.2815102-1-quic_abchauha@quicinc.comSigned-off-by:
-
Paolo Abeni authored
Eric Dumazet says: ==================== net: gro: cleanups and fast path refinement Current GRO stack has a 'fast path' for a subset of drivers, users of napi_frags_skb(). With TCP zerocopy/direct uses, header split at receive is becoming more important, and GRO fast path is disabled. This series makes GRO (a bit) more efficient for almost all use cases. ==================== Link: https://lore.kernel.org/r/20240301193740.3436871-1-edumazet@google.comSigned-off-by:
-
Eric Dumazet authored
In tcp_gro_complete() : Moving the skb->inner_transport_header setting allows the compiler to reuse the previously loaded value of skb->transport_header. Caching skb_shinfo() avoids duplications as well. In tcp4_gro_complete(), doing a single change on skb_shinfo(skb)->gso_type also generates better code. Signed-off-by:
-
Eric Dumazet authored
Currently the so-called GRO fast path is only enabled for napi_frags_skb() callers. After the prior patch, we no longer have to clear frag0 whenever we pulled bytes to skb->head. We therefore can initialize frag0 to skb->data so that GRO fast path can be used in the following additional cases: - Drivers using header split (populating skb->data with headers, and having payload in one or more page fragments). - Drivers not using any page frag (entire packet is in skb->data) Add a likely() in skb_gro_may_pull() to help the compiler to generate better code if possible. Signed-off-by:
-
Eric Dumazet authored
Change skb_gro_network_header() to accept a const sk_buff and to no longer check if frag0 is NULL or not. This allows to remove skb_gro_frag0_invalidate() which is seen in profiles when header-split is enabled. sk_buff parameter is constified for skb_gro_header_fast(), inet_gro_compute_pseudo() and ip6_gro_compute_pseudo(). Signed-off-by:
-
Eric Dumazet authored
skb_gro_header_hard() is renamed to skb_gro_may_pull() to match the convention used by common helpers like pskb_may_pull(). This means the condition is inverted: if (skb_gro_header_hard(skb, hlen)) slow_path(); becomes: if (!skb_gro_may_pull(skb, hlen)) slow_path(); Signed-off-by:
-
Paolo Abeni authored
says: ==================== MT7530 DSA Subdriver Improvements Act III This is the third patch series with the goal of simplifying the MT7530 DSA subdriver and improving support for MT7530, MT7531, and the switch on the MT7988 SoC. I have done a simple ping test to confirm basic communication on all switch ports on MCM and standalone MT7530, and MT7531 switch with this patch series applied. MT7621 Unielec, MCM MT7530: rgmii-only-gmac0-mt7621-unielec-u7621-06-16m.dtb gmac0-and-gmac1-mt7621-unielec-u7621-06-16m.dtb tftpboot 0x80008000 mips-uzImage.bin; tftpboot 0x83000000 mips-rootfs.cpio.uboot; tftpboot 0x83f00000 $dtb; bootm 0x80008000 0x83000000 0x83f00000 MT7622 Bananapi, MT7531: gmac0-and-gmac1-mt7622-bananapi-bpi-r64.dtb tftpboot 0x40000000 arm64-Image; tftpboot 0x45000000 arm64-rootfs.cpio.uboot; tftpboot 0x4a000000 $dtb; booti 0x40000000 0x45000000 0x4a000000 MT7623 Bananapi, standalone MT7530: rgmii-only-gmac0-mt7623n-bananapi-bpi-r2.dtb gmac0-and-gmac1-mt7623n-bananapi-bpi-r2.dtb tftpboot 0x80008000 arm-zImage; tftpboot 0x83000000 arm-rootfs.cpio.uboot; tftpboot 0x83f00000 $dtb; bootz 0x80008000 0x83000000 0x83f00000 This patch series is the continuation of the patch series linked below. https://lore.kernel.org/r/20230522121532.86610-1-arinc.unal@arinc9.comSigned-off-by:
Arınç ÜNAL <arinc.unal@arinc9.com> --- Changes in v3: - Patch 8 - Explain properly the behaviour of setting link down on all ports at setup. - Split the changes for simplifying the link settings operations out to another patch. - Link to v2: https://lore.kernel.org/r/20240216-for-netnext-mt7530-improvements-3-v2-0-094cae3ff23b@arinc9.com Changes in v2: - Patch 8 - Use a single mt7530_rmw() instead of two mt7530_clear() and mt7530_set() commands. - Link to v1: https://lore.kernel.org/r/20240208-for-netnext-mt7530-improvements-3-v1-0-d7c1cfd502ca@arinc9.com ==================== Link: https://lore.kernel.org/r/20240301-for-netnext-mt7530-improvements-3-v3-0-449f4f166454@arinc9.comSigned-off-by:
-
Arınç ÜNAL authored
The "MT7621 Giga Switch Programming Guide v0.3", "MT7531 Reference Manual for Development Board v1.0", and "MT7988A Wi-Fi 7 Generation Router Platform: Datasheet (Open Version) v0.1" documents show that these bits are enabled at reset: PMCR_IFG_XMIT(1) (not part of PMCR_LINK_SETTINGS_MASK) PMCR_MAC_MODE (not part of PMCR_LINK_SETTINGS_MASK) PMCR_TX_EN PMCR_RX_EN PMCR_BACKOFF_EN (not part of PMCR_LINK_SETTINGS_MASK) PMCR_BACKPR_EN (not part of PMCR_LINK_SETTINGS_MASK) PMCR_TX_FC_EN PMCR_RX_FC_EN These bits also don't exist on the MT7530_PMCR_P(6) register of the switch on the MT7988 SoC: PMCR_IFG_XMIT() PMCR_MAC_MODE PMCR_BACKOFF_EN PMCR_BACKPR_EN Remove the setting of the bits not part of PMCR_LINK_SETTINGS_MASK on phylink_mac_config as they're already set. The bit for setting the port on force mode is already done on mt7530_setup() and mt7531_setup_common(). So get rid of PMCR_FORCE_MODE_ID() which helped determine which bit to use for the switch model. Signed-off-by:
-
Arınç ÜNAL authored
port_enable and port_disable clears the link settings. Move that to mt7530_setup() and mt7531_setup_common() which set up the switches. This way, the link settings are cleared on all ports at setup, and then only once with phylink_mac_link_down() when a link goes down. Enable force mode at setup to apply the force part of the link settings. This ensures that disabled ports will have their link down. Suggested-by:
Vladimir Oltean <olteanv@gmail.com> Signed-off-by:
-
Arınç ÜNAL authored
The commit fae46308 ("net: dsa: mt753x: fix pcs conversion regression") fixes regression caused by cpu_port_config manually calling phylink operations. cpu_port_config was deemed useless and was removed. Therefore, put initialising PCS devices code back to its original order. Signed-off-by:
-
Arınç ÜNAL authored
There is no need for a separate function to call priv->info->mac_port_config(). Call it from mt753x_phylink_mac_config() instead and remove mt753x_mac_config(). Signed-off-by:
-
Arınç ÜNAL authored
priv->info->cpu_port_config() is used for MT7531 and the switch on the MT7988 SoC. It sets up the ports described as a CPU port earlier than the phylink code path would do. This function is useless as: - Configuring the MACs can be done from the phylink_mac_config code path instead. - All the link configuration it does on the CPU ports are later undone with the port_enable, phylink_mac_config, and then phylink_mac_link_up code path [1]. priv->p5_interface and priv->p6_interface were being used to prevent configuring the MACs from the phylink_mac_config code path. Remove them now that they hold no purpose. Remove priv->info->cpu_port_config(). On mt753x_phylink_mac_config, switch to if statements to simplify the code. Remove the overwriting of the speed and duplex interfaces for certain interface modes. Phylink already provides the speed and duplex variables with proper values. Phylink already sets the max speed of TRGMII to SPEED_1000. Add SPEED_2500 for PHY_INTERFACE_MODE_2500BASEX to where the speed and EEE bits are set instead. On the switch on the MT7988 SoC, PHY_INTERFACE_MODE_INTERNAL is being used to describe the interface mode of the 10G MAC, which is of port 6. On mt7988_cpu_port_config() PMCR_FORCE_SPEED_1000 was set via the PMCR_CPU_PORT_SETTING() mask. Add SPEED_10000 case to where the speed bits are set to cover this. No need to add it to where the EEE bits are set as the "MT7988A Wi-Fi 7 Generation Router Platform: Datasheet (Open Version) v0.1" document shows that these bits don't exist on the MT7530_PMCR_P(6) register. Remove the definition of PMCR_CPU_PORT_SETTING() now that it holds no purpose. Change mt753x_cpu_port_enable() to void now that there're no error cases left. Link: https://lore.kernel.org/netdev/ZHy2jQLesdYFMQtO@shell.armlinux.org.uk/ [1] Suggested-by:
Russell King (Oracle) <linux@armlinux.org.uk> Signed-off-by:
-
Arınç ÜNAL authored
Remove error returns on the cases where they are already handled with the function the mac_port_get_caps member in mt753x_table points to. mt7531_mac_config() is also called from mt7531_cpu_port_config() outside of phylink but the port and interface modes are already handled there. Change the functions and the mac_port_config function pointer to void now that there're no error returns anymore. Remove mt753x_is_mac_port() that used to help the said error returns. On mt7531_mac_config(), switch to if statements to simplify the code. Remove internal phy cases from mt753x_phylink_mac_config(), there is no need to check the interface mode as that's already handled with the function the mac_port_get_caps member in mt753x_table points to. Acked-by:
Daniel Golle <daniel@makrotopia.org> Tested-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by:
-
Arınç ÜNAL authored
According to the document MT7531 Reference Manual for Development Board v1.0, the SW_PHY_RST bit on the SYS_CTRL register doesn't exist for MT7531. This is likely why forcing link down on all ports is necessary for MT7531. Therefore, do not set SW_PHY_RST on mt7531_setup(). Signed-off-by:
-
Arınç ÜNAL authored
Setting this register related to interrupts is only needed for the MT7530 switch. Make an exclusive check to ensure this. Signed-off-by:
-
Arınç ÜNAL authored
For the switch on the MT7988 SoC, the mac_port_config member for ID_MT7988 in mt753x_table is not needed as the interfaces of all MACs are already handled on mt7988_mac_port_get_caps(). Therefore, remove the mac_port_config member from ID_MT7988 in mt753x_table. Before calling priv->info->mac_port_config(), if there's no mac_port_config member in mt753x_table, exit mt753x_mac_config() successfully. Remove calling priv->info->mac_port_config() from the sanity check as the sanity check requires a pointer to a mac_port_config function to be non-NULL. This will fail for MT7988 as mac_port_config won't be a member of its info table. Co-developed-by:
-
Paolo Abeni authored
Yunsheng Lin says: ==================== remove page frag implementation in vhost_net Currently there are three implementations for page frag: 1. mm/page_alloc.c: net stack seems to be using it in the rx part with 'struct page_frag_cache' and the main API being page_frag_alloc_align(). 2. net/core/sock.c: net stack seems to be using it in the tx part with 'struct page_frag' and the main API being skb_page_frag_refill(). 3. drivers/vhost/net.c: vhost seems to be using it to build xdp frame, and it's implementation seems to be a mix of the above two. This patchset tries to unfiy the page frag implementation a little bit by unifying gfp bit for order 3 page allocation and replacing page frag implementation in vhost.c with the one in page_alloc.c. After this patchset, we are not only able to unify the page frag implementation a little, but also able to have about 0.5% performance boost testing by using the vhost_net_test introduced in the last patch. Before this patchset: Performance counter stats for './vhost_net_test' (10 runs): 305325.78 msec task-clock # 1.738 CPUs utilized ( +- 0.12% ) 1048668 context-switches # 3.435 K/sec ( +- 0.00% ) 11 cpu-migrations # 0.036 /sec ( +- 17.64% ) 33 page-faults # 0.108 /sec ( +- 0.49% ) 244651819491 cycles # 0.801 GHz ( +- 0.43% ) (64) 64714638024 stalled-cycles-frontend # 26.45% frontend cycles idle ( +- 2.19% ) (67) 30774313491 stalled-cycles-backend # 12.58% backend cycles idle ( +- 7.68% ) (70) 201749748680 instructions # 0.82 insn per cycle # 0.32 stalled cycles per insn ( +- 0.41% ) (66.76%) 65494787909 branches # 214.508 M/sec ( +- 0.35% ) (64) 4284111313 branch-misses # 6.54% of all branches ( +- 0.45% ) (66) 175.699 +- 0.189 seconds time elapsed ( +- 0.11% ) After this patchset: Performance counter stats for './vhost_net_test' (10 runs): 303974.38 msec task-clock # 1.739 CPUs utilized ( +- 0.14% ) 1048807 context-switches # 3.450 K/sec ( +- 0.00% ) 14 cpu-migrations # 0.046 /sec ( +- 12.86% ) 33 page-faults # 0.109 /sec ( +- 0.46% ) 251289376347 cycles # 0.827 GHz ( +- 0.32% ) (60) 67885175415 stalled-cycles-frontend # 27.01% frontend cycles idle ( +- 0.48% ) (63) 27809282600 stalled-cycles-backend # 11.07% backend cycles idle ( +- 0.36% ) (71) 195543234672 instructions # 0.78 insn per cycle # 0.35 stalled cycles per insn ( +- 0.29% ) (69.04%) 62423183552 branches # 205.357 M/sec ( +- 0.48% ) (67) 4135666632 branch-misses # 6.63% of all branches ( +- 0.63% ) (67) 174.764 +- 0.214 seconds time elapsed ( +- 0.12% ) Changelog: V6: Add timeout for poll() and simplify some logic as suggested by Jason. V5: Address the comment from jason in vhost_net_test.c and the comment about leaving out the gfp change for page frag in sock.c as suggested by Paolo. V4: Resend based on latest net-next branch. V3: 1. Add __page_frag_alloc_align() which is passed with the align mask the original function expected as suggested by Alexander. 2. Drop patch 3 in v2 suggested by Alexander. 3. Reorder patch 4 & 5 in v2 suggested by Alexander. Note that placing this gfp flags handing for order 3 page in an inline function is not considered, as we may be able to unify the page_frag and page_frag_cache handling. V2: Change 'xor'd' to 'masked off', add vhost tx testing for vhost_net_test. V1: Fix some typo, drop RFC tag and rebase on latest net-next. ==================== Link: https://lore.kernel.org/r/20240228093013.8263-1-linyunsheng@huawei.comSigned-off-by: -
Yunsheng Lin authored
introduce vhost_net_test for both vhost_net tx and rx basing on virtio_test to test vhost_net changing in the kernel. Steps for vhost_net tx testing: 1. Prepare a out buf. 2. Kick the vhost_net to do tx processing. 3. Do the receiving in the tun side. 4. verify the data received by tun is correct. Steps for vhost_net rx testing: 1. Prepare a in buf. 2. Do the sending in the tun side. 3. Kick the vhost_net to do rx processing. 4. verify the data received by vhost_net is correct. Signed-off-by:
Yunsheng Lin <linyunsheng@huawei.com> Acked-by:
Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
-
Yunsheng Lin authored
The page frag in vhost_net_page_frag_refill() uses the 'struct page_frag' from skb_page_frag_refill(), but it's implementation is similar to page_frag_alloc_align() now. This patch removes vhost_net_page_frag_refill() by using 'struct page_frag_cache' instead of 'struct page_frag', and allocating frag using page_frag_alloc_align(). The added benefit is that not only unifying the page frag implementation a little, but also having about 0.5% performance boost testing by using the vhost_net_test introduced in the last patch. Signed-off-by:
Jason Wang <jasowang@redhat.com> Acked-by:
-
Yunsheng Lin authored
When draining a page_frag_cache, most user are doing the similar steps, so introduce an API to avoid code duplication. Signed-off-by:
Alexander Duyck <alexanderduyck@fb.com> Acked-by:
-
Yunsheng Lin authored
Currently there seems to be three page frag implementations which all try to allocate order 3 page, if that fails, it then fail back to allocate order 0 page, and each of them all allow order 3 page allocation to fail under certain condition by using specific gfp bits. The gfp bits for order 3 page allocation are different between different implementation, __GFP_NOMEMALLOC is or'd to forbid access to emergency reserves memory for __page_frag_cache_refill(), but it is not or'd in other implementions, __GFP_DIRECT_RECLAIM is masked off to avoid direct reclaim in vhost_net_page_frag_refill(), but it is not masked off in __page_frag_cache_refill(). This patch unifies the gfp bits used between different implementions by or'ing __GFP_NOMEMALLOC and masking off __GFP_DIRECT_RECLAIM for order 3 page allocation to avoid possible pressure for mm. Leave the gfp unifying for page frag implementation in sock.c for now as suggested by Paolo Abeni. Signed-off-by:
-
Yunsheng Lin authored
napi_alloc_frag_align() and netdev_alloc_frag_align() accept align as an argument, and they are thin wrappers around the __napi_alloc_frag_align() and __netdev_alloc_frag_align() APIs doing the alignment checking and align mask conversion, in order to call page_frag_alloc_align() directly. The intention here is to keep the alignment checking and the alignmask conversion in in-line wrapper to avoid those kind of operations during execution time since it can usually be handled during compile time. We are going to use page_frag_alloc_align() in vhost_net.c, it need the same kind of alignment checking and alignmask conversion, so split up page_frag_alloc_align into an inline wrapper doing the above operation, and add __page_frag_alloc_align() which is passed with the align mask the original function expected as suggested by Alexander. Signed-off-by:
-
Jiawen Wu authored
GPIO EOI is not set to clear interrupt status after handling the interrupt. It should be done in irq_chip->irq_ack, but this function is not called in handle_nested_irq(). So executing function txgbe_gpio_irq_ack() manually in txgbe_gpio_irq_handler(). Fixes: aefd0136 ("net: txgbe: use irq_domain for interrupt controller") Signed-off-by:
Jiawen Wu <jiawenwu@trustnetic.com> Link: https://lore.kernel.org/r/20240301092956.18544-2-jiawenwu@trustnetic.comSigned-off-by:
-
Jiawen Wu authored
The register of GPIO interrupt status is masked before MAC IRQ is enabled. This is because of hardware deficiency. So manually clear the interrupt status before using them. Otherwise, GPIO interrupts will never be reported again. There is a workaround for clearing interrupts to set GPIO EOI in txgbe_up_complete(). Fixes: aefd0136 ("net: txgbe: use irq_domain for interrupt controller") Signed-off-by:
-
Jakub Kicinski authored
Tony Nguyen says: ==================== Intel Wired LAN Driver Updates 2024-02-28 (ixgbe, igc, igb, e1000e, e100) This series contains updates to ixgbe, igc, igb, e1000e, and e100 drivers. Jon Maxwell makes module parameter values readable in sysfs for ixgbe, igb, and e100. Ernesto Castellotti adds support for 1000BASE-BX on ixgbe. Arnd Bergmann fixes build failure due to dependency issues for igc. Vitaly refactors error check to be more concise and prevent future issues on e1000e. v1: https://lore.kernel.org/netdev/20240229004135.741586-1-anthony.l.nguyen@intel.com/ ==================== Link: https://lore.kernel.org/r/20240301184806.2634508-1-anthony.l.nguyen@intel.comSigned-off-by:
-
Vitaly Lifshits authored
Add curly braces to avoid entering to an if statement where it is not always required in e1000_shutdown function. This improves code readability and might prevent non-deterministic behaviour in the future. Signed-off-by:
Vitaly Lifshits <vitaly.lifshits@intel.com> Tested-by:
Naama Meir <naamax.meir@linux.intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://lore.kernel.org/r/20240301184806.2634508-5-anthony.l.nguyen@intel.comSigned-off-by:
-
Arnd Bergmann authored
When IGC is built-in but LEDS_CLASS is a loadable module, there is a link failure: x86_64-linux-ld: drivers/net/ethernet/intel/igc/igc_leds.o: in function `igc_led_setup': igc_leds.c:(.text+0x75c): undefined reference to `devm_led_classdev_register_ext' Add another dependency that prevents this combination. Fixes: ea578703 ("igc: Add support for LEDs on i225/i226") Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Reviewed-by:
Kurt Kanzenbach <kurt@linutronix.de> Signed-off-by:
-
Ernesto Castellotti authored
Added support for 1000BASE-BX, i.e. Gigabit Ethernet over single strand of single-mode fiber. The initialization of a 1000BASE-BX SFP is the same as 1000BASE-SX/LX with the only difference that the Bit Rate Nominal Value must be checked to make sure it is a Gigabit Ethernet transceiver, as described by the SFF-8472 specification. This was tested with the FS.com SFP-GE-BX 1310/1490nm 10km transceiver: $ ethtool -m eth4 Identifier : 0x03 (SFP) Extended identifier : 0x04 (GBIC/SFP defined by 2-wire interface ID) Connector : 0x07 (LC) Transceiver codes : 0x00 0x00 0x00 0x40 0x00 0x00 0x00 0x00 0x00 Transceiver type : Ethernet: BASE-BX10 Encoding : 0x01 (8B/10B) BR, Nominal : 1300MBd Rate identifier : 0x00 (unspecified) Length (SMF,km) : 10km Length (SMF) : 10000m Length (50um) : 0m Length (62.5um) : 0m Length (Copper) : 0m Length (OM3) : 0m Laser wavelength : 1310nm Vendor name : FS Vendor OUI : 64:9d:99 Vendor PN : SFP-GE-BX Vendor rev : Option values : 0x20 0x0a Option : RX_LOS implemented Option : TX_FAULT implemented Option : Power level 3 requirement BR margin, max : 0% BR margin, min : 0% Vendor SN : S2202359108 Date code : 220307 Optical diagnostics support : Yes Laser bias current : 17.650 mA Laser output power : 0.2132 mW / -6.71 dBm Receiver signal average optical power : 0.2740 mW / -5.62 dBm Module temperature : 47.30 degrees C / 117.13 degrees F Module voltage : 3.2576 V Alarm/warning flags implemented : Yes Laser bias current high alarm : Off Laser bias current low alarm : Off Laser bias current high warning : Off Laser bias current low warning : Off Laser output power high alarm : Off Laser output power low alarm : Off Laser output power high warning : Off Laser output power low warning : Off Module temperature high alarm : Off Module temperature low alarm : Off Module temperature high warning : Off Module temperature low warning : Off Module voltage high alarm : Off Module voltage low alarm : Off Module voltage high warning : Off Module voltage low warning : Off Laser rx power high alarm : Off Laser rx power low alarm : Off Laser rx power high warning : Off Laser rx power low warning : Off Laser bias current high alarm threshold : 110.000 mA Laser bias current low alarm threshold : 1.000 mA Laser bias current high warning threshold : 100.000 mA Laser bias current low warning threshold : 1.000 mA Laser output power high alarm threshold : 0.7079 mW / -1.50 dBm Laser output power low alarm threshold : 0.0891 mW / -10.50 dBm Laser output power high warning threshold : 0.6310 mW / -2.00 dBm Laser output power low warning threshold : 0.1000 mW / -10.00 dBm Module temperature high alarm threshold : 90.00 degrees C / 194.00 degrees F Module temperature low alarm threshold : -45.00 degrees C / -49.00 degrees F Module temperature high warning threshold : 85.00 degrees C / 185.00 degrees F Module temperature low warning threshold : -40.00 degrees C / -40.00 degrees F Module voltage high alarm threshold : 3.7950 V Module voltage low alarm threshold : 2.8050 V Module voltage high warning threshold : 3.4650 V Module voltage low warning threshold : 3.1350 V Laser rx power high alarm threshold : 0.7079 mW / -1.50 dBm Laser rx power low alarm threshold : 0.0028 mW / -25.53 dBm Laser rx power high warning threshold : 0.6310 mW / -2.00 dBm Laser rx power low warning threshold : 0.0032 mW / -24.95 dBm Signed-off-by:Ernesto Castellotti <ernesto@castellotti.net> Reviewed-by:
Przemek Kitszel <przemyslaw.kitszel@intel.com> Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel) Signed-off-by:
-
Jon Maxwell authored
Linux users sometimes need an easy way to check current values of module parameters. For example the module may be manually reloaded with different parameters. Make these visible and readable in the /sys filesystem to allow that. But don't make the "debug" module parameter visible as debugging is enabled via ethtool msglvl. Signed-off-by:
Jon Maxwell <jmaxwell37@gmail.com> Reviewed-by:
Jacob Keller <jacob.e.keller@intel.com> Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel) Signed-off-by:
-
Eric Dumazet authored
Stephen Rothwell and kernel test robot reported that some arches (parisc, hexagon) and/or compilers would not like blamed commit. Lets make sure tcp_sock_write_rx group does not start with a hole. While we are at it, correct tcp_sock_write_tx CACHELINE_ASSERT_GROUP_SIZE() since after the blamed commit, we went to 105 bytes. Fixes: 99123622 ("tcp: remove some holes in struct tcp_sock") Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Reported-by:
kernel test robot <lkp@intel.com> Link: https://lore.kernel.org/netdev/20240301121108.5d39e4f9@canb.auug.org.au/ Closes: https://lore.kernel.org/oe-kbuild-all/202403011451.csPYOS3C-lkp@intel.com/Signed-off-by:
Simon Horman <horms@kernel.org> Tested-by: Simon Horman <horms@kernel.org> # build-tested Link: https://lore.kernel.org/r/20240301171945.2958176-1-edumazet@google.comSigned-off-by:
-
Pedro Tammela authored
Add the dependsOn test check for all the mirred blockcast tests. It will prevent the issue reported by LKFT which happens when an older iproute2 is used to run the current tdc. Tests are skipped if the dependsOn check fails. Reported-by:
Linux Kernel Functional Testing <lkft@linaro.org> Signed-off-by:
Pedro Tammela <pctammela@mojatatu.com> Link: https://lore.kernel.org/r/20240229143825.1373550-1-pctammela@mojatatu.comSigned-off-by:
-
