- 09 Nov, 2018 40 commits
-
-
David S. Miller authored
Jakub Kicinski says: ==================== nfp: abm: move code and improve parameter validation This set starts by separating Qdisc handling code into a new file. Next two patches allow early access to TLV-based capabilities during probe, previously the capabilities were parsed just before netdevs were registered, but its cleaner to do some basic validation earlier and avoid cleanup work. Next three patches improve RED's parameter validation. First we provide a more precise message about why offload failed (and move the parameter validation to a helper). Next we make sure we don't set the top bit in the 32 bit max RED threshold value. Because FW is treating the value as signed it reportedly causes slow downs (unnecessary queuing and marking) when top bit is set with recent firmwares. Last (and perhaps least importantly) we offload the harddrop parameter of the Qdisc. We don't plan to offload harddrop RED, but it seems prudent to make sure user didn't set that flag as device behaviour would have differed. ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Jakub Kicinski authored
RED Qdisc will now inform the drivers about the state of the harddrop flag. Refuse to offload in case harddrop is set. Signed-off-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by:
John Hurley <john.hurley@netronome.com> Reviewed-by:
Quentin Monnet <quentin.monnet@netronome.com> Signed-off-by:
-
Jakub Kicinski authored
To mirror software behaviour on offload more precisely inform the drivers about the state of the harddrop flag. Signed-off-by:
-
Jakub Kicinski authored
Turns out the threshold value is used in signed compares in the FW, so we should avoid setting the top bit. Signed-off-by:
-
Jakub Kicinski authored
Improve log messages printed when RED can't be offloaded because of Qdisc parameters. Signed-off-by:
-
Jakub Kicinski authored
In certain cases initialization logic which follows allocation of the vNIC structure may want to validate the capabilities of that vNIC. This is easy before vNIC is initialized for normal capabilities which are at fixed offsets in control memory, easy to locate and read, but poses a challenge if the capabilities are in form of TLVs. Parse the TLVs early on so other code can just access parsed info, instead of having to do the parsing by itself. Signed-off-by:
-
Jakub Kicinski authored
Move setting ctrl_bar pointer to the nfp_net_alloc function, to make sure we can parse capabilities early in the following patch. Signed-off-by:
-
Jakub Kicinski authored
The Qdisc offload code is logically separate, and we will soon do significant surgery on it to support more Qdiscs, so move it to a separate file. Signed-off-by:
-
Neal Cardwell authored
Recently, in commit ab408b6d ("tcp: switch tcp and sch_fq to new earliest departure time model"), the TCP BBR code switched to a new approach of using an explicit bbr_pacing_margin_percent for shaving a pacing rate "haircut", rather than the previous implict approach. Update an old comment to reflect the new approach. Signed-off-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
Yuchung Cheng <ycheng@google.com> Signed-off-by:
Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
David S. Miller authored
Michał Mirosław says: ==================== net: Use __vlan_hwaccel_*() helpers This series removes from networking core and driver code an assumption about how VLAN tag presence is stored in an skb. This will allow to free up overloading of VLAN.CFI bit to incidate tag's presence. ==================== Signed-off-by: -
Michał Mirosław authored
Signed-off-by:
Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
This removes assumption than vlan_tci != 0 when tag is present. Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Michał Mirosław authored
This removes assumptions about VLAN_TAG_PRESENT bit. Signed-off-by:
-
Michał Mirosław authored
Use __vlan_hwaccel_put_tag() to set vlan tag and proto fields. Signed-off-by:
-
Cong Wang authored
__skb_checksum_complete_head() and __skb_checksum_complete() are both declared in skbuff.h, they fit better in skbuff.c than datagram.c. Cc: Stefano Brivio <sbrivio@redhat.com> Signed-off-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
-
David S. Miller authored
Ivan Khoronzhuk says: ==================== net: ethernet: ti: cpsw: fix vlan mcast The cpsw holds separate mcast entires for vlan entries. At this moment driver adds only not vlan mcast addresses, omitting vlan/mcast entries. As result mcast for vlans doesn't work. It can be fixed by adding same mcast entries for every created vlan, but this patchseries uses more sophisticated way and allows to create mcast entries only for vlans that really require it. Generic functions from this series can be reused for fixing vlan and macvlan unicast. Simple example of ALE table before and after this series, having same mcast entries as for vlan 100 as for real device (reserved vlan 2), and one mcast address only for vlan 100 - 01:1b:19:00:00:00. <---- Before this patchset ----> vlan , vid = 2, untag_force = 0x5, reg_mcast = 0x5, mem_list = 0x5 mcast, vid = 2, addr = ff:ff:ff:ff:ff:ff, port_mask = 0x1 ucast, vid = 2, addr = 74:da:ea:47:7d:9d, persistant, port_num = 0x0 vlan , vid = 0, untag_force = 0x7, reg_mcast = 0x0, mem_list = 0x7 mcast, vid = 2, addr = 33:33:00:00:00:01, port_mask = 0x1 mcast, vid = 2, addr = 01:00:5e:00:00:01, port_mask = 0x1 vlan , vid = 1, untag_force = 0x3, reg_mcast = 0x3, mem_list = 0x3 mcast, vid = 1, addr = ff:ff:ff:ff:ff:ff, port_mask = 0x1 ucast, vid = 1, addr = 74:da:ea:47:7d:9c, persistant, port_num = 0x0 mcast, vid = 1, addr = 33:33:00:00:00:01, port_mask = 0x1 mcast, vid = 1, addr = 01:00:5e:00:00:01, port_mask = 0x1 mcast, vid = 2, addr = 01:80:c2:00:00:00, port_mask = 0x1 mcast, vid = 2, addr = 01:80:c2:00:00:03, port_mask = 0x1 mcast, vid = 2, addr = 01:80:c2:00:00:0e, port_mask = 0x1 mcast, vid = 1, addr = 01:80:c2:00:00:00, port_mask = 0x1 mcast, vid = 1, addr = 01:80:c2:00:00:03, port_mask = 0x1 mcast, vid = 1, addr = 01:80:c2:00:00:0e, port_mask = 0x1 mcast, vid = 2, addr = 33:33:ff:47:7d:9d, port_mask = 0x1 mcast, vid = 2, addr = 33:33:00:00:00:fb, port_mask = 0x1 mcast, vid = 2, addr = 33:33:00:01:00:03, port_mask = 0x1 mcast, vid = 1, addr = 33:33:ff:47:7d:9c, port_mask = 0x1 mcast, vid = 1, addr = 33:33:00:00:00:fb, port_mask = 0x1 mcast, vid = 1, addr = 33:33:00:01:00:03, port_mask = 0x1 mcast, vid = 1, addr = 01:00:5e:00:00:fb, port_mask = 0x1 mcast, vid = 1, addr = 01:00:5e:00:00:fc, port_mask = 0x1 vlan , vid = 100, untag_force = 0x0, reg_mcast = 0x5, mem_list = 0x5 ucast, vid = 100, addr = 74:da:ea:47:7d:9d, persistant, port_num = 0x0 mcast, vid = 100, addr = ff:ff:ff:ff:ff:ff, port_mask = 0x1 mcast, vid = 2, addr = 01:1b:19:00:00:00, port_mask = 0x1 ^^^ Here mcast entry (ptpl2), has to be added only for vlan 100 but added for reserved vlan 2...that's not enough. <---- After this patchset ----> vlan , vid = 2, untag_force = 0x5, reg_mcast = 0x5, mem_list = 0x5 mcast, vid = 2, addr = ff:ff:ff:ff:ff:ff, port_mask = 0x1 ucast, vid = 2, addr = 74:da:ea:47:7d:9d, persistant, port_num = 0x0 vlan , vid = 0, untag_force = 0x7, reg_mcast = 0x0, mem_list = 0x7 mcast, vid = 2, addr = 33:33:00:00:00:01, port_mask = 0x1 mcast, vid = 2, addr = 01:00:5e:00:00:01, port_mask = 0x1 vlan , vid = 1, untag_force = 0x3, reg_mcast = 0x3, mem_list = 0x3 mcast, vid = 1, addr = ff:ff:ff:ff:ff:ff, port_mask = 0x1 ucast, vid = 1, addr = 74:da:ea:47:7d:9c, persistant, port_num = 0x0 mcast, vid = 1, addr = 33:33:00:00:00:01, port_mask = 0x1 mcast, vid = 1, addr = 01:00:5e:00:00:01, port_mask = 0x1 mcast, vid = 2, addr = 01:80:c2:00:00:00, port_mask = 0x1 mcast, vid = 2, addr = 01:80:c2:00:00:03, port_mask = 0x1 mcast, vid = 2, addr = 01:80:c2:00:00:0e, port_mask = 0x1 mcast, vid = 1, addr = 01:80:c2:00:00:00, port_mask = 0x1 mcast, vid = 1, addr = 01:80:c2:00:00:03, port_mask = 0x1 mcast, vid = 1, addr = 01:80:c2:00:00:0e, port_mask = 0x1 mcast, vid = 2, addr = 33:33:ff:47:7d:9d, port_mask = 0x1 mcast, vid = 1, addr = 33:33:ff:47:7d:9c, port_mask = 0x1 mcast, vid = 2, addr = 33:33:00:00:00:fb, port_mask = 0x1 mcast, vid = 2, addr = 33:33:00:01:00:03, port_mask = 0x1 mcast, vid = 1, addr = 33:33:00:00:00:fb, port_mask = 0x1 mcast, vid = 1, addr = 33:33:00:01:00:03, port_mask = 0x1 vlan , vid = 100, untag_force = 0x0, reg_mcast = 0x5, mem_list = 0x5 ucast, vid = 100, addr = 74:da:ea:47:7d:9d, persistant, port_num = 0x0 mcast, vid = 100, addr = ff:ff:ff:ff:ff:ff, port_mask = 0x1 mcast, vid = 100, addr = 33:33:00:00:00:01, port_mask = 0x1 mcast, vid = 100, addr = 01:00:5e:00:00:01, port_mask = 0x1 mcast, vid = 100, addr = 33:33:ff:47:7d:9d, port_mask = 0x1 mcast, vid = 100, addr = 01:80:c2:00:00:00, port_mask = 0x1 mcast, vid = 100, addr = 01:80:c2:00:00:03, port_mask = 0x1 mcast, vid = 100, addr = 01:80:c2:00:00:0e, port_mask = 0x1 mcast, vid = 100, addr = 33:33:00:00:00:fb, port_mask = 0x1 mcast, vid = 100, addr = 33:33:00:01:00:03, port_mask = 0x1 mcast, vid = 100, addr = 01:1b:19:00:00:00, port_mask = 0x1 ^^^ Here mcast entry (ptpl2), is added only for vlan 100 as it should be. Based on net-next/master v2..v1: net: ethernet: ti: cpsw: fix vlan mcast - removed limit for legacy switch cpsw mode ==================== Signed-off-by: -
Ivan Khoronzhuk authored
The vlan configuration is not restored after interface donw/up sequence (if dual-emac - both interfaces). Tested on am572x EVM. Steps to check: ~# ip link add link eth1 name eth1.100 type vlan id 100 ~# ifconfig eth0 down ~# ifconfig eth1 down Try to remove vid and observe warning: ~# ip link del eth1.100 [ 739.526757] net eth1: removing vlanid 100 from vlan filter [ 739.533322] failed to kill vid 0081/100 for device eth1 This patch fixes it, restoring only vlan ALE entries and all other unicast/multicast entries are restored by system calling rx_mode ndo. Reviewed-by:
Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by:
Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org> Signed-off-by:
-
Ivan Khoronzhuk authored
At this moment, mcast addresses are added for real device only (reserved vlans for dual-emac mode), even if a mcast address was added for some vlan only, thus ALE doesn't have corresponding vlan mcast entries after vlan socket joined multicast group. So ALE drops vlan frames with mcast addresses intended for vlans and potentially can receive mcast frames for base ndev. That's not correct. So, fix it by creating only vlan/mcast entries as requested. Patch doesn't use any additional lists and is based on device mc address list and cpsw ALE table entries. Signed-off-by:
-
Ivan Khoronzhuk authored
It's redundancy for the drivers to hold the list of vlans when absolutely the same list exists in vlan core. In most cases it's needed only to traverse the vlan devices, their vids and sync some settings with h/w, so add API to simplify this. At least some of these drivers also can benefit: grep "for_each.*vid" -r drivers/net/ethernet/ drivers/net/ethernet/hisilicon/hns3/hns3_enet.c: drivers/net/ethernet/synopsys/dwc-xlgmac-hw.c: drivers/net/ethernet/qlogic/qlge/qlge_main.c: drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c: drivers/net/ethernet/via/via-rhine.c: drivers/net/ethernet/via/via-velocity.c: drivers/net/ethernet/intel/igb/igb_main.c: drivers/net/ethernet/intel/ice/ice_main.c: drivers/net/ethernet/intel/e1000/e1000_main.c: drivers/net/ethernet/intel/i40e/i40e_main.c: drivers/net/ethernet/intel/e1000e/netdev.c: drivers/net/ethernet/intel/igbvf/netdev.c: drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c: drivers/net/ethernet/intel/ixgb/ixgb_main.c: drivers/net/ethernet/intel/ixgbe/ixgbe_main.c: drivers/net/ethernet/amd/xgbe/xgbe-dev.c: drivers/net/ethernet/emulex/benet/be_main.c: drivers/net/ethernet/neterion/vxge/vxge-main.c: drivers/net/ethernet/adaptec/starfire.c: drivers/net/ethernet/brocade/bna/bnad.c: Reviewed-by:
-
Ivan Khoronzhuk authored
In order to avoid all table update, and only remove or add new address, the auxiliary function exists, named __hw_addr_sync_dev(). It allows end driver do nothing when nothing changed and add/rm when concrete address is firstly added or lastly removed. But it doesn't include cases when an address of real device or vlan was reused by other vlans or vlan/macval devices. For handaling events when address was reused/unreused the patch adds new auxiliary routine - __hw_addr_ref_sync_dev(). It allows to do nothing when nothing was changed and do updates only for an address being added/reused/deleted/unreused. Thus, clone address changes for vlans can be mirrored in the table. The function is exclusive with __hw_addr_sync_dev(). It's responsibility of the end driver to identify address vlan device, if it needs so. Signed-off-by:
-
Edward Cree authored
As added in 3e59020a ("net: bql: add __netdev_tx_sent_queue()"), which see for performance rationale. Signed-off-by:
Edward Cree <ecree@solarflare.com> Reviewed-by:
-
David S. Miller authored
Michał Mirosław says: ==================== net: Remove VLAN_TAG_PRESENT from drivers This series removes VLAN_TAG_PRESENT use from network drivers in preparation to removing its special meaning. ==================== Signed-off-by: -
Michał Mirosław authored
Reviewed-by:
Claudiu Manoil <claudiu.manoil@nxp.com> Signed-off-by:
-
Michał Mirosław authored
This is a minimal change to allow removing of VLAN_TAG_PRESENT. It leaves OVS unable to use CFI bit, as fixing this would need a deeper surgery involving userspace interface. Signed-off-by:
-
Michał Mirosław authored
This just removes VLAN_TAG_PRESENT use. VLAN TCI=0 special meaning is deeply embedded in the driver code and so is left as is. Signed-off-by:
-
Michał Mirosław authored
Signed-off-by:
-
Ilias Apalodimas authored
return -ENOMEM directly instead of assigning it in a variable Signed-off-by:
Ilias Apalodimas <ilias.apalodimas@linaro.org> Signed-off-by:
-
Ilias Apalodimas authored
Current driver dynamically allocates an skb and maps it as DMA Rx buffer. In order to prepare for upcoming XDP changes, let's introduce a different allocation scheme. Buffers are allocated dynamically and mapped into hardware. During the Rx operation the driver uses build_skb() to produce the necessary buffers for the network stack. This change increases performance ~15% on 64b packets with smmu disabled and ~5% with smmu enabled Signed-off-by:
-
Stefan Wahren authored
Interferences on the SPI line could distort the response of available buffer space. So at least we should check that the response doesn't exceed the maximum available buffer space. In error case increase a new error counter and retry it later. This behavior avoids buffer errors in the QCA7000, which results in an unnecessary chip reset including packet loss. Signed-off-by:
Stefan Wahren <stefan.wahren@i2se.com> Signed-off-by:
-
David Barmann authored
When setting the SO_MARK socket option, if the mark changes, the dst needs to be reset so that a new route lookup is performed. This fixes the case where an application wants to change routing by setting a new sk_mark. If this is done after some packets have already been sent, the dst is cached and has no effect. Signed-off-by:
David Barmann <david.barmann@stackpath.com> Reviewed-by:
-
David S. Miller authored
Julian Wiedmann says: ==================== s390/qeth: updates 2018-11-08 please apply the following qeth patches to net-next. The first patch allows one more device type to query the FW for a MAC address, the others are all basically just removal of duplicated or unused code. ==================== Signed-off-by: -
Julian Wiedmann authored
qeth_l3_setup_netdev() checks if the hsuid attribute is set on the qeth device, and propagates it to the net_device. In the past this was needed to pick up any hsuid that was set before allocation of the net_device. With commit d3d1b205 ("s390/qeth: allocate netdevice early") this is no longer necessary, qeth_l3_dev_hsuid_store() always stores the hsuid straight into dev->perm_addr. Signed-off-by:
Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by:
-
Julian Wiedmann authored
If the CREATE ADDR sent by qeth_l3_iqd_read_initial_mac() fails, its callback sets a random MAC address on the net_device. The error then propagates back, and qeth_l3_setup_netdev() bails out without registering the net_device. Any subsequent call to qeth_l3_setup_netdev() will then attempt a fresh CREATE ADDR which either 1) also fails, or 2) sets a proper MAC address on the net_device. Consequently, the net_device will never be registered with a random MAC and we can drop the fallback code. Signed-off-by:
-
Julian Wiedmann authored
qeth_l3_send_ipa_arp_cmd() is merely a wrapper around qeth_send_control_data() now. So push the length adjustment into QETH_SETASS_BASE_LEN, and remove the wrapper. While at it, also remove some redundant 0-initializations. qeth_send_setassparms() requires that callers prepare their command parameters, so that they can be copied into the parameter area in one go. Skip the indirection, and just let callers set up the command themselves. Signed-off-by:
-
Julian Wiedmann authored
Call qeth_prepare_ipa_cmd() during setup of a new IPA cmd buffer, so that it is used for all commands. Thus ARP and SNMP requests don't have to do their own initialization. This will now also set the proper MPC protocol version for SNMP requests on L2 devices. Signed-off-by:
-
