Linux Ethernet Bridging - Apr 2022 - [Bridge] [PATCH net-next] net: bridge: offload BR_HAIRPIN_MODE, BR_ISOLATED, BR_MULTICAST_TO_UNICAST

Add BR_HAIRPIN_MODE, BR_ISOLATED and BR_MULTICAST_TO_UNICAST port flags to
BR_PORT_FLAGS_HW_OFFLOAD so that switchdev drivers which have an offloaded
data plane have a chance to reject these bridge port flags if they don't
support them yet.

It makes the code path go through the
SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS driver handlers, which return
-EINVAL for everything they don't recognize.

For drivers that don't catch SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS at
all, switchdev will return -EOPNOTSUPP for those which is then ignored, but
those are in the minority.

Signed-off-by: Ar?n? ?NAL <arinc.unal at arinc9.com>
---
Let me know if this is netdev/net material instead.

Commit log is heavily quoted from Vladimir Oltean <olteanv at gmail.com>.

Ar?n?
---
 net/bridge/br_switchdev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c
index 8cc44c367231..81400e0b26ac 100644
--- a/net/bridge/br_switchdev.c
+++ b/net/bridge/br_switchdev.c
@@ -72,7 +72,8 @@ bool nbp_switchdev_allowed_egress(const struct net_bridge_port
*p,
 
 /* Flags that can be offloaded to hardware */
 #define BR_PORT_FLAGS_HW_OFFLOAD (BR_LEARNING | BR_FLOOD | \
-				  BR_MCAST_FLOOD | BR_BCAST_FLOOD | BR_PORT_LOCKED)
+				  BR_MCAST_FLOOD | BR_BCAST_FLOOD | BR_PORT_LOCKED | \
+				  BR_HAIRPIN_MODE | BR_ISOLATED | BR_MULTICAST_TO_UNICAST)
 
 int br_switchdev_set_port_flag(struct net_bridge_port *p,
 			       unsigned long flags,
-- 
2.25.1

On Sun, Apr 10, 2022 at 04:42:27PM +0300, Ar?n? ?NAL
wrote:
> Add BR_HAIRPIN_MODE, BR_ISOLATED and BR_MULTICAST_TO_UNICAST port flags to
> BR_PORT_FLAGS_HW_OFFLOAD so that switchdev drivers which have an offloaded
> data plane have a chance to reject these bridge port flags if they
don't
> support them yet.
> 
> It makes the code path go through the
> SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS driver handlers, which return
> -EINVAL for everything they don't recognize.
> 
> For drivers that don't catch SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS at
> all, switchdev will return -EOPNOTSUPP for those which is then ignored, but
> those are in the minority.
> 
> Signed-off-by: Ar?n? ?NAL <arinc.unal at arinc9.com>
Reviewed-by: Ido Schimmel <idosch at nvidia.com>
> ---
> Let me know if this is netdev/net material instead.
I prefer net-next

On Sun, Apr 10, 2022 at 04:42:27PM +0300, Ar?n? ?NAL
wrote:
> Add BR_HAIRPIN_MODE, BR_ISOLATED and BR_MULTICAST_TO_UNICAST port flags to
> BR_PORT_FLAGS_HW_OFFLOAD so that switchdev drivers which have an offloaded
> data plane have a chance to reject these bridge port flags if they
don't
> support them yet.
> 
> It makes the code path go through the
> SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS driver handlers, which return
> -EINVAL for everything they don't recognize.
> 
> For drivers that don't catch SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS at
> all, switchdev will return -EOPNOTSUPP for those which is then ignored, but
> those are in the minority.
> 
> Signed-off-by: Ar?n? ?NAL <arinc.unal at arinc9.com>
> ---
Reviewed-by: Vladimir Oltean <olteanv at gmail.com>
> Let me know if this is netdev/net material instead.
The thing with targeting this against "net" is that we've seen the
introduction of BR_PORT_LOCKED fairly recently which has tainted the
BR_PORT_FLAGS_HW_OFFLOAD macro. So backporting would conflict very
quickly down the path. Sure that isn't a decisive problem, but I don't
think it's worth the extra trouble of preparing special patches for the
"stable" trees, and having those diverge from the current master.
> 
> Commit log is heavily quoted from Vladimir Oltean <olteanv at
gmail.com>.
> 
> Ar?n?
> ---
>  net/bridge/br_switchdev.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c
> index 8cc44c367231..81400e0b26ac 100644
> --- a/net/bridge/br_switchdev.c
> +++ b/net/bridge/br_switchdev.c
> @@ -72,7 +72,8 @@ bool nbp_switchdev_allowed_egress(const struct
net_bridge_port *p,
>  
>  /* Flags that can be offloaded to hardware */
>  #define BR_PORT_FLAGS_HW_OFFLOAD (BR_LEARNING | BR_FLOOD | \
> -				  BR_MCAST_FLOOD | BR_BCAST_FLOOD | BR_PORT_LOCKED)
> +				  BR_MCAST_FLOOD | BR_BCAST_FLOOD | BR_PORT_LOCKED | \
> +				  BR_HAIRPIN_MODE | BR_ISOLATED | BR_MULTICAST_TO_UNICAST)
>  
>  int br_switchdev_set_port_flag(struct net_bridge_port *p,
>  			       unsigned long flags,
> -- 
> 2.25.1
>

Hello:

This patch was applied to netdev/net-next.git (master)
by Jakub Kicinski <kuba at kernel.org>:

On Sun, 10 Apr 2022 16:42:27 +0300 you wrote:
> Add BR_HAIRPIN_MODE, BR_ISOLATED and BR_MULTICAST_TO_UNICAST port flags to
> BR_PORT_FLAGS_HW_OFFLOAD so that switchdev drivers which have an offloaded
> data plane have a chance to reject these bridge port flags if they
don't
> support them yet.
> 
> It makes the code path go through the
> SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS driver handlers, which return
> -EINVAL for everything they don't recognize.
> 
> [...]
Here is the summary with links:
  - [net-next] net: bridge: offload BR_HAIRPIN_MODE, BR_ISOLATED,
BR_MULTICAST_TO_UNICAST
    https://git.kernel.org/netdev/net-next/c/c3976a3f8445

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html