| Message ID | 20210212151600.3357121-1-olteanv@gmail.com |
|---|---|
| Headers | show |
| Series | Cleanup in brport flags switchdev offload for DSA | expand |
On 12/02/2021 17:15, Vladimir Oltean wrote: > From: Vladimir Oltean <vladimir.oltean@nxp.com> > > When a struct switchdev_attr is notified through switchdev, there is no > way to report informational messages, unlike for struct switchdev_obj. > > Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> > Reviewed-by: Ido Schimmel <idosch@nvidia.com> > Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> > --- > Changes in v5: > Rebased on top of AM65 CPSW driver merge. > > Changes in v4: > None. > > Changes in v3: > None. > > Changes in v2: > Patch is new. > > .../ethernet/marvell/prestera/prestera_switchdev.c | 3 ++- > .../net/ethernet/mellanox/mlxsw/spectrum_switchdev.c | 3 ++- > drivers/net/ethernet/mscc/ocelot_net.c | 3 ++- > drivers/net/ethernet/ti/am65-cpsw-switchdev.c | 3 ++- > drivers/net/ethernet/ti/cpsw_switchdev.c | 3 ++- > include/net/switchdev.h | 6 ++++-- > net/dsa/slave.c | 3 ++- > net/switchdev/switchdev.c | 11 ++++++++--- > 8 files changed, 24 insertions(+), 11 deletions(-) > Reviewed-by: Nikolay Aleksandrov <nikolay@nvidia.com>
On 12/02/2021 17:15, Vladimir Oltean wrote: > From: Vladimir Oltean <vladimir.oltean@nxp.com> > > This switchdev attribute offers a counterproductive API for a driver > writer, because although br_switchdev_set_port_flag gets passed a > "flags" and a "mask", those are passed piecemeal to the driver, so while > the PRE_BRIDGE_FLAGS listener knows what changed because it has the > "mask", the BRIDGE_FLAGS listener doesn't, because it only has the final > value. But certain drivers can offload only certain combinations of > settings, like for example they cannot change unicast flooding > independently of multicast flooding - they must be both on or both off. > The way the information is passed to switchdev makes drivers not > expressive enough, and unable to reject this request ahead of time, in > the PRE_BRIDGE_FLAGS notifier, so they are forced to reject it during > the deferred BRIDGE_FLAGS attribute, where the rejection is currently > ignored. > > This patch also changes drivers to make use of the "mask" field for edge > detection when possible. > > Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> > --- > Changes in v5: > Rebased on top of AM65 CPSW driver. > > Changes in v4: > Patch is new. > > .../marvell/prestera/prestera_switchdev.c | 23 +++++---- > .../mellanox/mlxsw/spectrum_switchdev.c | 50 +++++++++++-------- > drivers/net/ethernet/rocker/rocker_main.c | 10 ++-- > drivers/net/ethernet/ti/am65-cpsw-switchdev.c | 24 +++++---- > drivers/net/ethernet/ti/cpsw_switchdev.c | 24 +++++---- > drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 34 ++++++++----- > include/net/switchdev.h | 7 ++- > net/bridge/br_switchdev.c | 6 +-- > net/dsa/dsa_priv.h | 6 ++- > net/dsa/port.c | 34 +++++++------ > 10 files changed, 129 insertions(+), 89 deletions(-) > Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
On 2/12/2021 7:15 AM, Vladimir Oltean wrote: > From: Vladimir Oltean <vladimir.oltean@nxp.com> > > For the netlink interface, propagate errors through extack rather than > simply printing them to the console. For the sysfs interface, we still > print to the console, but at least that's one layer higher than in > switchdev, which also allows us to silently ignore the offloading of > flags if that is ever needed in the future. > > Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> I suppose the slight "loss" of information on which port failed the operation is okay since it was implied by either the path (sysfs) or the bridge command.
Hello: This series was applied to netdev/net-next.git (refs/heads/master): On Fri, 12 Feb 2021 17:15:50 +0200 you wrote: > From: Vladimir Oltean <vladimir.oltean@nxp.com> > > The initial goal of this series was to have better support for > standalone ports mode on the DSA drivers like ocelot/felix and sja1105. > This turned out to require some API adjustments in both directions: > to the information presented to and by the switchdev notifier, and to > the API presented to the switch drivers by the DSA layer. > > [...] Here is the summary with links: - [v5,net-next,01/10] net: switchdev: propagate extack to port attributes https://git.kernel.org/netdev/net-next/c/4c08c586ff29 - [v5,net-next,02/10] net: bridge: offload all port flags at once in br_setport https://git.kernel.org/netdev/net-next/c/304ae3bf1c1a - [v5,net-next,03/10] net: bridge: don't print in br_switchdev_set_port_flag https://git.kernel.org/netdev/net-next/c/078bbb851ea6 - [v5,net-next,04/10] net: dsa: configure better brport flags when ports leave the bridge https://git.kernel.org/netdev/net-next/c/5e38c15856e9 - [v5,net-next,05/10] net: switchdev: pass flags and mask to both {PRE_,}BRIDGE_FLAGS attributes https://git.kernel.org/netdev/net-next/c/e18f4c18ab5b - [v5,net-next,06/10] net: dsa: act as passthrough for bridge port flags https://git.kernel.org/netdev/net-next/c/a8b659e7ff75 - [v5,net-next,07/10] net: dsa: felix: restore multicast flood to CPU when NPI tagger reinitializes https://git.kernel.org/netdev/net-next/c/6edb9e8d451e - [v5,net-next,08/10] net: mscc: ocelot: use separate flooding PGID for broadcast https://git.kernel.org/netdev/net-next/c/b360d94f1b86 - [v5,net-next,09/10] net: mscc: ocelot: offload bridge port flags to device https://git.kernel.org/netdev/net-next/c/421741ea5672 - [v5,net-next,10/10] net: dsa: sja1105: offload bridge port flags to device https://git.kernel.org/netdev/net-next/c/4d9423549501 You are awesome, thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html
From: Vladimir Oltean <vladimir.oltean@nxp.com> The initial goal of this series was to have better support for standalone ports mode on the DSA drivers like ocelot/felix and sja1105. This turned out to require some API adjustments in both directions: to the information presented to and by the switchdev notifier, and to the API presented to the switch drivers by the DSA layer. Vladimir Oltean (10): net: switchdev: propagate extack to port attributes net: bridge: offload all port flags at once in br_setport net: bridge: don't print in br_switchdev_set_port_flag net: dsa: configure better brport flags when ports leave the bridge net: switchdev: pass flags and mask to both {PRE_,}BRIDGE_FLAGS attributes net: dsa: act as passthrough for bridge port flags net: dsa: felix: restore multicast flood to CPU when NPI tagger reinitializes net: mscc: ocelot: use separate flooding PGID for broadcast net: mscc: ocelot: offload bridge port flags to device net: dsa: sja1105: offload bridge port flags to device drivers/net/dsa/b53/b53_common.c | 91 ++++--- drivers/net/dsa/b53/b53_priv.h | 2 - drivers/net/dsa/mv88e6xxx/chip.c | 163 ++++++++++--- drivers/net/dsa/mv88e6xxx/chip.h | 6 +- drivers/net/dsa/mv88e6xxx/port.c | 52 ++-- drivers/net/dsa/mv88e6xxx/port.h | 19 +- drivers/net/dsa/ocelot/felix.c | 25 ++ drivers/net/dsa/sja1105/sja1105.h | 2 + drivers/net/dsa/sja1105/sja1105_main.c | 222 +++++++++++++++++- drivers/net/dsa/sja1105/sja1105_spi.c | 6 + .../marvell/prestera/prestera_switchdev.c | 26 +- .../mellanox/mlxsw/spectrum_switchdev.c | 53 +++-- drivers/net/ethernet/mscc/ocelot.c | 100 +++++++- drivers/net/ethernet/mscc/ocelot_net.c | 52 +++- drivers/net/ethernet/rocker/rocker_main.c | 10 +- drivers/net/ethernet/ti/am65-cpsw-switchdev.c | 27 ++- drivers/net/ethernet/ti/cpsw_switchdev.c | 27 ++- drivers/staging/fsl-dpaa2/ethsw/ethsw.c | 34 ++- include/net/dsa.h | 10 +- include/net/switchdev.h | 13 +- include/soc/mscc/ocelot.h | 20 +- net/bridge/br_netlink.c | 116 +++------ net/bridge/br_private.h | 6 +- net/bridge/br_switchdev.c | 23 +- net/bridge/br_sysfs_if.c | 7 +- net/dsa/dsa_priv.h | 11 +- net/dsa/port.c | 76 ++++-- net/dsa/slave.c | 10 +- net/switchdev/switchdev.c | 11 +- 29 files changed, 889 insertions(+), 331 deletions(-)