Ido Schimmel
2021-Jul-20 11:24 UTC
[Bridge] [PATCH v4 net-next 00/15] Allow forwarding for the software bridge data path to be offloaded to capable devices
On Mon, Jul 19, 2021 at 12:44:19AM +0300, Vladimir Oltean wrote:> drivers/net/dsa/mv88e6xxx/chip.c | 78 +++- > .../ethernet/freescale/dpaa2/dpaa2-switch.c | 69 +++- > .../ethernet/marvell/prestera/prestera_main.c | 99 +++-- > .../marvell/prestera/prestera_switchdev.c | 42 ++- > .../marvell/prestera/prestera_switchdev.h | 7 +- > .../net/ethernet/mellanox/mlxsw/spectrum.c | 347 ++++++++++++------ > .../net/ethernet/mellanox/mlxsw/spectrum.h | 4 + > .../mellanox/mlxsw/spectrum_switchdev.c | 28 +- > .../microchip/sparx5/sparx5_switchdev.c | 48 ++- > drivers/net/ethernet/mscc/ocelot_net.c | 115 ++++-- > drivers/net/ethernet/rocker/rocker.h | 9 +- > drivers/net/ethernet/rocker/rocker_main.c | 34 +- > drivers/net/ethernet/rocker/rocker_ofdpa.c | 42 ++- > drivers/net/ethernet/ti/am65-cpsw-nuss.c | 34 +- > drivers/net/ethernet/ti/am65-cpsw-switchdev.c | 14 +- > drivers/net/ethernet/ti/am65-cpsw-switchdev.h | 3 + > drivers/net/ethernet/ti/cpsw_new.c | 32 +- > drivers/net/ethernet/ti/cpsw_switchdev.c | 4 +- > drivers/net/ethernet/ti/cpsw_switchdev.h | 3 + > include/linux/if_bridge.h | 63 ++-- > include/net/dsa.h | 21 ++ > net/bridge/br_fdb.c | 1 - > net/bridge/br_forward.c | 9 + > net/bridge/br_if.c | 11 +- > net/bridge/br_mdb.c | 1 - > net/bridge/br_private.h | 84 ++++- > net/bridge/br_switchdev.c | 287 +++++++++++++-- > net/bridge/br_vlan.c | 11 +- > net/dsa/dsa2.c | 4 + > net/dsa/dsa_priv.h | 6 + > net/dsa/port.c | 192 +++++++--- > net/dsa/tag_dsa.c | 52 ++- > 32 files changed, 1406 insertions(+), 348 deletions(-)Too many things are squashed into this one patchset. It needs to be split. According to the title, the patchset is focused on improving performance, but there are no performance numbers that I could see and most of the patches deal with the replay stuff instead. The TX forwarding offload in mv88e6xxx is not related to the replay stuff and should be added in a separate patchset. This can be done by first adding the switchdev_bridge_port_offload() / switchdev_bridge_port_unoffload() APIs that only take care of setting / unsetting the hardware domain for the bridge port. Then, in a different patchset, these APIs can be augmented with a parameter for the replay stuff. It should be easier to review that way and require less unnecessary surgeries in drivers that do not require the added functionality. Thanks
Vladimir Oltean
2021-Jul-20 13:20 UTC
[Bridge] [PATCH v4 net-next 00/15] Allow forwarding for the software bridge data path to be offloaded to capable devices
On Tue, Jul 20, 2021 at 02:24:29PM +0300, Ido Schimmel wrote:> Too many things are squashed into this one patchset. It needs to be > split. > > The TX forwarding offload in mv88e6xxx is not related to the replay > stuff and should be added in a separate patchset. This can be done by > first adding the switchdev_bridge_port_offload() / > switchdev_bridge_port_unoffload() APIs that only take care of setting / > unsetting the hardware domain for the bridge port. Then, in a different > patchset, these APIs can be augmented with a parameter for the replay > stuff. It should be easier to review that way and require less > unnecessary surgeries in drivers that do not require the added > functionality.Fair point. I will submit patches 1-10 and 11-15 separately.> According to the title, the patchset is focused on improving > performance, but there are no performance numbers that I could see and > most of the patches deal with the replay stuff instead.Maybe, but the truth is that it is not really the performance improvement that I care about. The performance quote is from Tobias' original cover letter, which I took as-is. I can build a synthetic test for multicasting on 10 mv88e6xxx ports or something like that, or maybe Tobias can provide a more relevant example out of Westermo's use cases. But it would be silly if this patchset's acceptance would depend on the numbers. This is one of those cases where completely different interests led me and Tobias to the the same solution. I don't want to bore you to death with details, but for some switches (DSA or otherwise), being able to send bridge packets as they are (data plane packets) instead of what they aren't (control plane packets) is a matter of functionality and not performance. Such switches only use control plane packets for link-local packet traps, and sending/receiving a control packet is expensive. For this class of switches (some may call them "dumb", but whatever), this patch series makes the difference between supporting and not supporting local IP termination through a VLAN-aware bridge, bridging with a foreign interface, bridging with software upper interfaces like LAG, etc.