kernel.git/net/dsa/dsa.c, branch linux-5.11.y

net: dsa: use net core stats64 handling

2020-11-10T01:50:27Z

Use netdev->tstats instead of a member of dsa_slave_priv for storing a pointer to the per-cpu counters. This allows us to use core functionality for statistics handling. Reviewed-by: Florian Fainelli Tested-by: Vladimir Oltean Signed-off-by: Heiner Kallweit Signed-off-by: Jakub Kicinski

net: dsa: Add devlink port regions support to DSA

2020-10-04T21:38:53Z

Allow DSA drivers to make use of devlink port regions, via simple wrappers. Reviewed-by: Vladimir Oltean Tested-by: Vladimir Oltean Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller

net: dsa: Call dsa_untag_bridge_pvid() from dsa_switch_rcv()

2020-10-02T20:36:07Z

When a DSA switch driver needs to call dsa_untag_bridge_pvid(), it can set dsa_switch::untag_brige_pvid to indicate this is necessary. This is a pre-requisite to making sure that we are always calling dsa_untag_bridge_pvid() after eth_type_trans() has been called. Signed-off-by: Florian Fainelli Reviewed-by: Vladimir Oltean Signed-off-by: David S. Miller

net: dsa: Add devlink regions support to DSA

2020-09-19T01:17:45Z

Allow DSA drivers to make use of devlink regions, via simple wrappers. Reviewed-by: Florian Fainelli Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller

net: dsa: Add helper to convert from devlink to ds

2020-09-19T01:17:45Z

Given a devlink instance, return the dsa switch it is associated to. Reviewed-by: Florian Fainelli Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller

net: dsa: introduce a dsa_port_from_netdev public helper

2020-05-08T00:31:57Z

As its implementation shows, this is synonimous with calling dsa_slave_dev_check followed by dsa_slave_to_port, so it is quite simple already and provides functionality which is already there. However there is now a need for these functions outside dsa_priv.h, for example in drivers that perform mirroring and redirection through tc-flower offloads (they are given raw access to the flow_cls_offload structure), where they need to call this function on act->dev. But simply exporting dsa_slave_to_port would make it non-inline and would result in an extra function call in the hotpath, as can be seen for example in sja1105: Before: 000006dc : { 6dc: e92d4ff0 push {r4, r5, r6, r7, r8, r9, sl, fp, lr} 6e0: e1a04000 mov r4, r0 6e4: e591958c ldr r9, [r1, #1420] ; 0x58c <- Inline dsa_slave_to_port 6e8: e1a05001 mov r5, r1 6ec: e24dd004 sub sp, sp, #4 u16 tx_vid = dsa_8021q_tx_vid(dp->ds, dp->index); 6f0: e1c901d8 ldrd r0, [r9, #24] 6f4: ebfffffe bl 0 6f4: R_ARM_CALL dsa_8021q_tx_vid u8 pcp = netdev_txq_to_tc(netdev, queue_mapping); 6f8: e1d416b0 ldrh r1, [r4, #96] ; 0x60 u16 tx_vid = dsa_8021q_tx_vid(dp->ds, dp->index); 6fc: e1a08000 mov r8, r0 After: 000006e4 : { 6e4: e92d4ff0 push {r4, r5, r6, r7, r8, r9, sl, fp, lr} 6e8: e1a04000 mov r4, r0 6ec: e24dd004 sub sp, sp, #4 struct dsa_port *dp = dsa_slave_to_port(netdev); 6f0: e1a00001 mov r0, r1 { 6f4: e1a05001 mov r5, r1 struct dsa_port *dp = dsa_slave_to_port(netdev); 6f8: ebfffffe bl 0 6f8: R_ARM_CALL dsa_slave_to_port 6fc: e1a09000 mov r9, r0 u16 tx_vid = dsa_8021q_tx_vid(dp->ds, dp->index); 700: e1c001d8 ldrd r0, [r0, #24] 704: ebfffffe bl 0 704: R_ARM_CALL dsa_8021q_tx_vid Because we want to avoid possible performance regressions, introduce this new function which is designed to be public. Suggested-by: Vivien Didelot Signed-off-by: Vladimir Oltean Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller

net: dsa: add GRO support via gro_cells

2020-04-23T19:32:13Z

gro_cells lib is used by different encapsulating netdevices, such as geneve, macsec, vxlan etc. to speed up decapsulated traffic processing. CPU tag is a sort of "encapsulation", and we can use the same mechs to greatly improve overall DSA performance. skbs are passed to the GRO layer after removing CPU tags, so we don't need any new packet offload types as it was firstly proposed by me in the first GRO-over-DSA variant [1]. The size of struct gro_cells is sizeof(void *), so hot struct dsa_slave_priv becomes only 4/8 bytes bigger, and all critical fields remain in one 32-byte cacheline. The other positive side effect is that drivers for network devices that can be shipped as CPU ports of DSA-driven switches can now use napi_gro_frags() to pass skbs to kernel. Packets built that way are completely non-linear and are likely being dropped without GRO. This was tested on to-be-mainlined-soon Ethernet driver that uses napi_gro_frags(), and the overall performance was on par with the variant from [1], sometimes even better due to minimal overhead. net.core.gro_normal_batch tuning may help to push it to the limit on particular setups and platforms. iperf3 IPoE VLAN NAT TCP forwarding (port1.218 -> port0) setup on 1.2 GHz MIPS board: 5.7-rc2 baseline: [ID] Interval Transfer Bitrate Retr [ 5] 0.00-120.01 sec 9.00 GBytes 644 Mbits/sec 413 sender [ 5] 0.00-120.00 sec 8.99 GBytes 644 Mbits/sec receiver Iface RX packets TX packets eth0 7097731 7097702 port0 426050 6671829 port1 6671681 425862 port1.218 6671677 425851 With this patch: [ID] Interval Transfer Bitrate Retr [ 5] 0.00-120.01 sec 12.2 GBytes 870 Mbits/sec 122 sender [ 5] 0.00-120.00 sec 12.2 GBytes 870 Mbits/sec receiver Iface RX packets TX packets eth0 9474792 9474777 port0 455200 353288 port1 9019592 455035 port1.218 353144 455024 v2: - Add some performance examples in the commit message; - No functional changes. [1] https://lore.kernel.org/netdev/20191230143028.27313-1-alobakin@dlink.ru/ Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller

net: dsa: Simplify 'dsa_tag_protocol_to_str()'

2020-03-30T17:54:57Z

There is no point in preparing the module name in a buffer. The format string can be passed diectly to 'request_module()'. This axes a few lines of code and cleans a few things: - max len for a driver name is MODULE_NAME_LEN wich is ~ 60 chars, not 128. It would be down-sized in 'request_module()' - we should pass the total size of the buffer to 'snprintf()', not the size minus 1 Signed-off-by: Christophe JAILLET Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller

net: dsa: Add support for devlink resources

2019-11-06T02:09:45Z

Add wrappers around the devlink resource API, so that DSA drivers can register and unregister devlink resources. Signed-off-by: Andrew Lunn Acked-by: Jiri Pirko Signed-off-by: David S. Miller

net: dsa: Add support for devlink device parameters

2019-10-28T23:21:02Z

Add plumbing to allow DSA drivers to register parameters with devlink. To keep with the abstraction, the DSA drivers pass the ds structure to these helpers, and the DSA core then translates that to the devlink structure associated to the device. Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller