- 08 Jan, 2018 40 commits
-
-
Fuyun Liang authored
When setting MTU, actually what we do is configuring the max frame size for the hardware. ETH_HLEN、ETH_FCS_LEN and VLAN_HLEN must need to be considered. And the frame size which is less than the default value should not be set to the hardware. Because in the hardware, the the max frame size not only controls the RX packet size, but also controls the TX packet size. the RX packets whose size are greater than the setting value will be dropped. This patch fixes the bug setting a error max frame size to hardware. Fixes: 46a3df9f ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by:
Fuyun Liang <liangfuyun1@huawei.com> Signed-off-by:
Peng Li <lipeng321@huawei.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Fuyun Liang authored
commit a9c782822166 ("net: hns3: add support for set_pauseparam") adds set_pauseparam support for ethtool cmd, but forgets to update fc_mode_last_time when PFC mode is disabled in hclge_cfg_pauseparam(). The wrong fc_mode_last_time will be used to update flow control mode when lldpad has been running. As a result, when using the ethtool command "-a", user will get a wrong pause parameter. This patch adds the fc_mode_last_time update when PFC mode is disabled. Fixes: a9c782822166 ("net: hns3: add support for set_pauseparam") Signed-off-by: -
Jian Shen authored
The result of tqp statistics query was read with an error position, fix it according to the user manual. Fixes: 46a3df9f ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by:
Jian Shen <shenjian15@huawei.com> Signed-off-by:
-
Jian Shen authored
Add packet statistics of netdev for ethtool -S, in order to show the statistics data for current net device. Remove update_stats() calling because it has been completed in hns3_get_netdev_stats(). Signed-off-by:
-
Jian Shen authored
The member "stats_size" of struct hns3_stats is useless, remove it and fix the macro definition which has uses this struct. Signed-off-by:
-
Jian Shen authored
The member "stats_offset" was designed to indicate the offset of each member of struct ring_stats in struct hns3_enet_ring, but forgot to add the offset of the member in struct ring_stats. Fixes: 496d03e9 ("net: hns3: Add Ethtool support to HNS3 driver") Signed-off-by:
-
Jian Shen authored
An error loop index was used while querying statistics data of tqps, which may cause call trace. Fixes: 496d03e9 ("net: hns3: Add Ethtool support to HNS3 driver") Signed-off-by:
-
Jian Shen authored
The dropped tx/rx packets number of each tqp should also be counted into the total drop tx/rx packets numbers. Fixes: 76ad4f0e ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC") Signed-off-by:
-
Jian Shen authored
Update the HNS3_NIC_STATE_DOWN bit when NIC state changes. When NIC is down, mask the packet statistics for querying with ifconfig command. It's a common practice. Signed-off-by:
-
Jian Shen authored
It takes more than 200 query response messages between driver and IMP, while updating the packet statistics. It's too heavy for IMP to update it per second. Extend the update period of packet statistics data from 1 second to 300 seconds(if too long, the statistics may overflow). As a result, we need to update it while querying with ifconfig tool to keep the statistics data fresh. Signed-off-by:
-
Jian Shen authored
The igu_rx_err_pkt indicates the same error with mac_rx_fcs_err_pkt_num, so remove it. Fixes: 46a3df9f ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by:
-
Jian Shen authored
Fix spelling error "overrsize" --> "oversize". Fixes: 46a3df9f ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by:
-
Jian Shen authored
Some members of packet statistics are named in different styles. This patch unifies them with new internal name rules, the main modification are below: trans --> tx rcv --> rx rcb_q%d_tx --> txq#%d rcb_q%d_rx --> rxq#%d sw_err_cnt(tx side) --> tx_dropped sw_err_cnt(rx side) --> rx_dropped pkts --> packets tx_err_cnt --> errors rx_err_cnt --> errors Signed-off-by:
-
Jian Shen authored
Rxvlan offload status can only be changed by PF. Initialize the value of NETIF_F_HW_VLAN_CTAG_RX bit of hw_features for VFS to false, make sure user can't be able to change it. Signed-off-by:
-
Jian Shen authored
This patch adds vlan filter enable switch to support ethtool -K ethX rx-vlan-filter on/off. Signed-off-by:
-
David S. Miller authored
Subash Abhinov Kasiviswanathan says: ==================== net: qualcomm: rmnet: Enable csum offloads This series introduces the MAPv4 packet format for checksum offload plus some other minor changes. Patches 1-3 are cleanups. Patch 4 renames the ingress format to data format so that all data formats can be configured using this going forward. Patch 5 uses the pacing helper to improve TCP transmit performance. Patch 6-9 defines the the MAPv4 for checksum offload for RX and TX. A new header and trailer format are used as part of MAPv4. For RX checksum offload, only the 1's complement of the IP payload portion is computed by hardware. The meta data from RX header is used to verify the checksum field in the packet. Note that the IP packet and its field itself is not modified by hardware. This gives metadata to help with the RX checksum. For TX, the required metadata is filled up so hardware can compute the checksum. Patch 10 enables GSO on rmnet devices v1->v2: Fix sparse errors reported by kbuild test robot v2->v3: Update the commit message for Patch 5 based on Eric's comments ==================== Signed-off-by: -
Subash Abhinov Kasiviswanathan authored
Real devices may support scatter gather(SG), so enable SG on rmnet devices to use GSO. GSO reduces CPU cycles by 20% for a rate of 146Mpbs for a single stream TCP connection. Signed-off-by:
Subash Abhinov Kasiviswanathan <subashab@codeaurora.org> Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
TX checksum offload applies to TCP / UDP packets which are not fragmented using the MAPv4 checksum trailer. The following needs to be done to have checksum computed in hardware - 1. Set the checksum start offset and inset offset. 2. Set the csum_enabled bit 3. Compute and set 1's complement of partial checksum field in transport header. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
When using the MAPv4 packet format in conjunction with MAP commands, a dummy DL checksum trailer will be appended to the packet. Before this packet is sent out as an ACK, the DL checksum trailer needs to be removed. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
When using the MAPv4 packet format, receive checksum offload can be enabled in hardware. The checksum computation over pseudo header is not offloaded but the rest of the checksum computation over the payload is offloaded. This applies only for TCP / UDP packets which are not fragmented. rmnet validates the TCP/UDP checksum for the packet using the checksum from the checksum trailer added to the packet by hardware. The validation performed is as following - 1. Perform 1's complement over the checksum value from the trailer 2. Compute 1's complement checksum over IPv4 / IPv6 header and subtracts it from the value from step 1 3. Computes 1's complement checksum over IPv4 / IPv6 pseudo header and adds it to the value from step 2 4. Subtracts the checksum value from the TCP / UDP header from the value from step 3. 5. Compares the value from step 4 to the checksum value from the TCP / UDP header. 6. If the comparison in step 5 succeeds, CHECKSUM_UNNECESSARY is set and the packet is passed on to network stack. If there is a failure, then the packet is passed on as such without modifying the ip_summed field. The checksum field is also checked for UDP checksum 0 as per RFC 768 and for unexpected TCP checksum of 0. If checksum offload is disabled when using MAPv4 packet format in receive path, the packet is queued as is to network stack without the validations above. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
The MAPv4 packet format adds support for RX / TX checksum offload. For a bi-directional UDP stream at a rate of 570 / 146 Mbps, roughly 10% CPU cycles are saved. For receive path, there is a checksum trailer appended to the end of the MAP packet. The valid field indicates if hardware has computed the checksum. csum_start_offset indicates the offset from the start of the IP header from which hardware has computed checksum. csum_length is the number of bytes over which the checksum was computed and the resulting value is csum_value. In the transmit path, a header is appended between the end of the MAP header and the start of the IP packet. csum_start_offset is the offset in bytes from which hardware will compute the checksum if the csum_enabled bit is set. udp_ip4_ind indicates if the checksum value of 0 is valid or not. csum_insert_offset is the offset from the csum_start_offset where hardware will insert the computed checksum. The use of this additional packet format for checksum offload is explained in subsequent patches. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
The real device over which the rmnet devices are installed also aggregate multiple IP packets and sends them as a single large aggregate frame to the hardware. This causes degraded throughput for TCP TX due to bufferbloat. To overcome this problem, pacing shift value of 8 is set using the sk_pacing_shift_update() helper. This value was determined based on experiments with a single stream TCP TX using iperf for a duration of 30s. Pacing shift | Observed data rate (Mbps) 10 | 9 9 | 140 8 | 146 (Max link rate) Signed-off-by: -
Subash Abhinov Kasiviswanathan authored
This is done so that we can use this field for both ingress and egress flags. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
rmnet_map_demultiplex() is only declared but not defined anywhere, so remove it. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
rmnet devices cannot have a mux id of 255. This is validated when assigning the mux id to the rmnet devices. As a result, checking for mux id 255 does not apply in egress path. Signed-off-by:
-
Subash Abhinov Kasiviswanathan authored
We already check the headroom once in rmnet_map_egress_handler(), so this is not needed. Signed-off-by:
-
David S. Miller authored
Ido Schimmel says: ==================== ipv6: Align nexthop behaviour with IPv4 This set tries to eliminate some differences between IPv4's and IPv6's treatment of nexthops. These differences are most likely a side effect of IPv6's data structures (specifically 'rt6_info') that incorporate both the route and the nexthop and the late addition of ECMP support in commit 51ebd318 ("ipv6: add support of equal cost multipath (ECMP)"). IPv4 and IPv6 do not react the same to certain netdev events. For example, upon carrier change affected IPv4 nexthops are marked using the RTNH_F_LINKDOWN flag and the nexthop group is rebalanced accordingly. IPv6 on the other hand, does nothing which forces us to perform a carrier check during route lookup and dump. This makes it difficult to introduce features such as non-equal-cost multipath that are built on top of this set [1]. In addition, when a netdev is put administratively down IPv4 nexthops are marked using the RTNH_F_DEAD flag, whereas IPv6 simply flushes all the routes using these nexthops. To be consistent with IPv4, multipath routes should only be flushed when all nexthops in the group are considered dead. The first 12 patches introduce non-functional changes that store the RTNH_F_DEAD and RTNH_F_LINKDOWN flags in IPv6 routes based on netdev events, in a similar fashion to IPv4. This allows us to remove the carrier check performed during route lookup and dump. The next three patches make sure we only flush a multipath route when all of its nexthops are dead. Last three patches add test cases for IPv4/IPv6 FIB. These verify that both address families react similarly to netdev events. Finally, this series also serves as a good first step towards David Ahern's goal of treating nexthops as standalone objects [2], as it makes the code more in line with IPv4 where the nexthop and the nexthop group are separate objects from the route itself. 1. https://github.com/idosch/linux/tree/ipv6-nexthops 2. http://vger.kernel.org/netconf2017_files/nexthop-objects.pdf Changes since RFC (feedback from David Ahern): * Remove redundant declaration of rt6_ifdown() in patch 4 and adjust comment referencing it accordingly * Drop patch to flush multipath routes upon NETDEV_UNREGISTER. Reword cover letter accordingly * Use a temporary variable to make code more readable in patch 15 ==================== Signed-off-by:
-
Ido Schimmel authored
Check that IPv4 and IPv6 react the same when the carrier of a netdev is toggled. Local routes should not be affected by this, whereas unicast routes should. Signed-off-by:
Ido Schimmel <idosch@mellanox.com> Acked-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
Ido Schimmel authored
Check that IPv4 and IPv6 react the same when a netdev is being put administratively down. Signed-off-by:
-
Ido Schimmel authored
Add test cases to check that IPv4 and IPv6 react to a netdev being unregistered as expected. Signed-off-by:
-
Ido Schimmel authored
By default, IPv6 deletes nexthops from a multipath route when the nexthop device is put administratively down. This differs from IPv4 where the nexthops are kept, but marked with the RTNH_F_DEAD flag. A multipath route is flushed when all of its nexthops become dead. Align IPv6 with IPv4 and have it conform to the same guidelines. In case the multipath route needs to be flushed, its siblings are flushed one by one. Otherwise, the nexthops are marked with the appropriate flags and the tree walker is instructed to skip all the siblings. As explained in previous patches, care is taken to update the sernum of the affected tree nodes, so as to prevent the use of wrong dst entries. Signed-off-by:
-
Ido Schimmel authored
The next patch is going to allow dead routes to remain in the FIB tree in certain situations. When this happens we need to be sure to bump the sernum of the nodes where these are stored so that potential copies cached in sockets are invalidated. The function that performs this update assumes the table lock is not taken when it is invoked, but that will not be the case when it is invoked by the tree walker. Have the function assume the lock is taken and make the single caller take the lock itself. Signed-off-by:
-
Ido Schimmel authored
We are going to allow dead routes to stay in the FIB tree (e.g., when they are part of a multipath route, directly connected route with no carrier) and revive them when their nexthop device gains carrier or when it is put administratively up. This is equivalent to the addition of the route to the FIB tree and we should therefore take care of updating the sernum of all the parent nodes of the node where the route is stored. Otherwise, we risk sockets caching and using sub-optimal dst entries. Export the function that performs the above, so that it could be invoked from fib6_ifup() later on. Signed-off-by:
-
Ido Schimmel authored
As explained in previous patch, fib6_ifdown() needs to consider the state of all the sibling routes when a multipath route is traversed. This is done by evaluating all the siblings when the first sibling in a multipath route is traversed. If the multipath route does not need to be flushed (e.g., not all siblings are dead), then we should just skip the multipath route as our work is done. Have the tree walker jump to the last sibling when it is determined that the multipath route needs to be skipped. Signed-off-by:
-
Ido Schimmel authored
When routes that are a part of a multipath route are evaluated by fib6_ifdown() in response to NETDEV_DOWN and NETDEV_UNREGISTER events the state of their sibling routes is not considered. This will change in subsequent patches in order to align IPv6 with IPv4's behavior. For example, when the last sibling in a multipath route becomes dead, the entire multipath route needs to be removed. To prevent the tree walker from re-evaluating all the sibling routes each time, we can simply evaluate them once - when the first sibling is traversed. If we determine the entire multipath route needs to be removed, then the 'should_flush' bit is set in all the siblings, which will cause the walker to flush them when it traverses them. Signed-off-by:
-
Ido Schimmel authored
Up until now the RTNH_F_DEAD flag was only reported in route dump when the 'ignore_routes_with_linkdown' sysctl was set. This is expected as dead routes were flushed otherwise. The reliance on this sysctl is going to be removed, so we need to report the flag regardless of the sysctl's value. Signed-off-by:
-
Ido Schimmel authored
Currently, dead routes are only present in the routing tables in case the 'ignore_routes_with_linkdown' sysctl is set. Otherwise, they are flushed. Subsequent patches are going to remove the reliance on this sysctl and make IPv6 more consistent with IPv4. Before this is done, we need to make sure dead routes are skipped during route lookup, so as to not cause packet loss. Signed-off-by:
-
Ido Schimmel authored
Similar to previous patch, there is no need to check for the carrier of the nexthop device when dumping the route and we can instead check for the presence of the RTNH_F_LINKDOWN flag. Signed-off-by:
-
Ido Schimmel authored
Now that the RTNH_F_LINKDOWN flag is set in nexthops, we can avoid the need to dereference the nexthop device and check its carrier and instead check for the presence of the flag. Signed-off-by:
-
Ido Schimmel authored
It is valid to install routes with a nexthop device that does not have a carrier, so we need to make sure they're marked accordingly. As explained in the previous patch, host and anycast routes are never marked with the 'linkdown' flag. Note that reject routes are unaffected, as these use the loopback device which always has a carrier. Signed-off-by:
-
