| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <82317367-CCF4-4E21-84B1-911F8388E3E7@bamaicloud.com>
Date: Sun, 11 May 2025 22:34:17 +0800
From: Tonghao Zhang <tonghao@...aicloud.com>
To: Jay Vosburgh <jv@...sburgh.net>
Cc: netdev@...r.kernel.org,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
Simon Horman <horms@...nel.org>,
Jonathan Corbet <corbet@....net>,
Andrew Lunn <andrew+netdev@...n.ch>
Subject: Re: [PATCH net-next 3/4] net: bonding: send peer notify when failure
recovery
> 2025年5月10日 下午9:01，Jay Vosburgh <jv@...sburgh.net> 写道：
>
> tonghao@...aicloud.com wrote:
>
>> From: Tonghao Zhang <tonghao@...aicloud.com>
>>
>> While hardware failures in NICs, optical transceivers, or switches
>> are unavoidable, rapid system recovery can be achieved post-restoration.
>> For example, triggering immediate ARP/ND packet transmission upon
>> LACP failure recovery enables the system to swiftly resume normal
>> operations, thereby minimizing service downtime.
>
> I think this comment needs to be prefaced with something that
> explains that this logic is for the "no stack" architecture. It don't
> need the entire blurb about what that is, though.
Ok, Update the commit message in v2.
>
>> Cc: Jay Vosburgh <jv@...sburgh.net>
>> Cc: "David S. Miller" <davem@...emloft.net>
>> Cc: Eric Dumazet <edumazet@...gle.com>
>> Cc: Jakub Kicinski <kuba@...nel.org>
>> Cc: Paolo Abeni <pabeni@...hat.com>
>> Cc: Simon Horman <horms@...nel.org>
>> Cc: Jonathan Corbet <corbet@....net>
>> Cc: Andrew Lunn <andrew+netdev@...n.ch>
>> Signed-off-by: Tonghao Zhang <tonghao@...aicloud.com>
>> ---
>> drivers/net/bonding/bond_3ad.c | 14 ++++++++++++++
>> 1 file changed, 14 insertions(+)
>>
>> diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
>> index c6807e473ab7..6577ce54d115 100644
>> --- a/drivers/net/bonding/bond_3ad.c
>> +++ b/drivers/net/bonding/bond_3ad.c
>> @@ -982,6 +982,19 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker)
>> return 0;
>> }
>>
>> +static void ad_peer_notif_send(struct port *port)
>> +{
>> + if (!port->aggregator->is_active)
>> + return;
>> +
>> + struct bonding *bond = port->slave->bond;
>> + if (bond->params.broadcast_neighbor && rtnl_trylock()) {
>> + bond->send_peer_notif = bond->params.num_peer_notif *
>> + max(1, bond->params.peer_notif_delay);
>> + rtnl_unlock();
>> + }
>> +}
>> +
>
> I'm not a fan of the function name, as this doesn't actually
> send any notifications. Perhaps "ad_cond_set_peer_notif"? I.e.,
> conditionally set peer notifications on?
>
>> /**
>> * ad_mux_machine - handle a port's mux state machine
>> * @port: the port we're looking at
>> @@ -1164,6 +1177,7 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
>> port->actor_oper_port_state |= LACP_STATE_COLLECTING;
>> port->actor_oper_port_state |= LACP_STATE_DISTRIBUTING;
>> port->actor_oper_port_state |= LACP_STATE_SYNCHRONIZATION;
>> + ad_peer_notif_send(port);
>> ad_enable_collecting_distributing(port,
>> update_slave_arr);
>> port->ntt = true;
>
> This is in the AD_MUX_COLLECTING_DISTRIBUTING case, I think you
> need another one of these in the AD_MUX_DISTRIBUTING case a few lines
> further down to handle the situation when coupled_control is disabled.
pretty good, moving the ad_cond_set_peer_notif to ad_enable_collecting_distributing makes more sense.
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index c6807e473ab7..d1c2d416ac87 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -982,6 +982,17 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker)
return 0;
}
+static void ad_cond_set_peer_notif(struct port *port)
+{
+ struct bonding *bond = port->slave->bond;
+
+ if (bond->params.broadcast_neighbor && rtnl_trylock()) {
+ bond->send_peer_notif = bond->params.num_peer_notif *
+ max(1, bond->params.peer_notif_delay);
+ rtnl_unlock();
+ }
+}
+
/**
* ad_mux_machine - handle a port's mux state machine
* @port: the port we're looking at
@@ -2061,6 +2072,8 @@ static void ad_enable_collecting_distributing(struct port *port,
__enable_port(port);
/* Slave array needs update */
*update_slave_arr = true;
+ /* Should notify peers if possible */
+ ad_cond_set_peer_notif(port);
}
}
By the way, bond_should_notify_peers should check params.broadcast_neighbor, because curr_active_slave may be NULL. In my test environment, ip li set slave-ethx down and when up again, bond_change_active_slave will set bond->curr_active_slave in lacp mode.
static bool bond_should_notify_peers(struct bonding *bond)
{
struct bond_up_slave *slaves;
struct slave *slave = NULL;
if (BOND_MODE(bond) == BOND_MODE_8023AD) {
if (!bond->params.broadcast_neighbor)
return false;
slaves = rtnl_dereference(bond->usable_slaves);
if (!slaves || !READ_ONCE(slaves->count))
return false;
} else {
slave = rcu_dereference_rtnl(bond->curr_active_slave);
if (!save || test_bit(__LINK_STATE_LINKWATCH_PENDING,
&slave->dev->state))
return false;
}
if (!bond->send_peer_notif ||
bond->send_peer_notif %
max(1, bond->params.peer_notif_delay) != 0 ||
!netif_carrier_ok(bond->dev))
return false;
netdev_dbg(bond->dev, "bond_should_notify_peers: slave %s\n",
slave ? slave->dev->name : "all");
return true;
}
>
> -J
>
>> --
>> 2.34.1
>>
>
> ---
> -Jay Vosburgh, jv@...sburgh.net
Powered by blists - more mailing lists