For FT2000+/64 devices,
when four virtual machines share the same physical network interface,
DROP will occur due to the single core CPU performance problem.

ip_check_defrag and macvlan_process_broadcast is on the same CPU.
When the MACVLAN PORT increases, the CPU usage reaches more than 90%.
bc_queue > bc_queue_len_used (default 1000), causing DROP.

Signed-off-by: zhangxiangqian <zhangxiangqian@kylinos.cn>
---
 drivers/net/macvlan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 1080d6e..dd3f35e 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -361,7 +361,7 @@ static void macvlan_broadcast_enqueue(struct macvlan_port *port,
 	}
 	spin_unlock(&port->bc_queue.lock);
 
-	schedule_work(&port->bc_work);
+	queue_work(system_unbound_wq, &port->bc_work);
 
 	if (err)
 		goto free_nskb;
-- 
2.7.4

Hello:

This patch was applied to netdev/net.git (master)
by David S. Miller <davem@davemloft.net>:

On Thu, 13 Oct 2022 15:41:12 +0800 you wrote:
> For FT2000+/64 devices,
> when four virtual machines share the same physical network interface,
> DROP will occur due to the single core CPU performance problem.
> 
> ip_check_defrag and macvlan_process_broadcast is on the same CPU.
> When the MACVLAN PORT increases, the CPU usage reaches more than 90%.
> bc_queue > bc_queue_len_used (default 1000), causing DROP.
> 
> [...]

Here is the summary with links:
  - net: macvlan: change schedule system_wq to system_unbound_wq
    https://git.kernel.org/netdev/net/c/3d6642eac74d

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