* [dpdk-dev] [PATCH] drivers/net/bonding: fix bug for lacp negotiation failed
@ 2020-05-19 7:53 luyicai
0 siblings, 0 replies; 5+ messages in thread
From: luyicai @ 2020-05-19 7:53 UTC (permalink / raw)
To: dev
Cc: chas3, zhangliang, Zhoujingbin (Robin, Russell Lab),
chenchanghu, Lilijun (Jerry)
When two host is connected directly without any devices like switch,
and also enable dedicated tx/rx queues on bonding devices slaves,
rx_machine_update would recieving partner lacp negotiation packets,
which partner's port mac filled with zeros. So in this situation,
it would never go rx_machine branch with correct mac!
Thus bond mode 4 will negotiation failed.
Signed-off-by: Lu Yicai <luyicai@huawei.com>
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
index b77a37ddb..2002ec04a 100644
--- a/drivers/net/bonding/rte_eth_bond_8023ad.c
+++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
@@ -798,7 +798,8 @@ rx_machine_update(struct bond_dev_private *internals, uint16_t slave_id,
RTE_ASSERT(lacp->lacpdu.subtype == SLOW_SUBTYPE_LACP);
partner = &lacp->lacpdu.partner;
- if (rte_is_same_ether_addr(&partner->port_params.system,
+ if (rte_is_zero_ether_addr(&partner->port_params.system) ||
+ rte_is_same_ether_addr(&partner->port_params.system,
&internals->mode4.mac_addr)) {
/* This LACP frame is sending to the bonding port
* so pass it to rx_machine.
--
2.19.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH] drivers/net/bonding: fix bug for lacp negotiation failed
2020-07-10 21:07 ` Thomas Monjalon
@ 2020-07-11 1:42 ` Wei Hu (Xavier)
0 siblings, 0 replies; 5+ messages in thread
From: Wei Hu (Xavier) @ 2020-07-11 1:42 UTC (permalink / raw)
To: Thomas Monjalon, luyicai
Cc: dev, chas3, zhangliang, chenchanghu, jerry.lilijun, haifeng.lin,
guohongzhi1, ferruh.yigit, matan, rasland, Wei Hu (Xavier)
Hi, Thomas Monjalon
The patch V2 has been sent by luyicai, and V2 is good to me.
Thanks, Xavier
On 2020/7/11 5:07, Thomas Monjalon wrote:
> Still no review.
>
> Please give your real name in the Signed-off so it can be merged.
>
> 24/06/2020 12:43, Thomas Monjalon:
>> Any review please?
>>
>> 19/05/2020 13:18, luyicai:
>>> When two host is connected directly without any devices like switch,
>>> and also enable dedicated tx/rx queues on bonding devices slaves,
>>> rx_machine_update would recieving partner lacp negotiation packets,
>>> which partner's port mac filled with zeros. So in this situation,
>>> it would never go rx_machine branch with correct mac!
>>> Thus bond mode 4 will negotiation failed.
>>>
>>> Signed-off-by: luyicai <luyicai@huawei.com>
>>> ---
>>> drivers/net/bonding/rte_eth_bond_8023ad.c | 3 ++-
>>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
>>> index b77a37d..2002ec0 100644
>>> --- a/drivers/net/bonding/rte_eth_bond_8023ad.c
>>> +++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
>>> @@ -798,7 +798,8 @@
>>> RTE_ASSERT(lacp->lacpdu.subtype == SLOW_SUBTYPE_LACP);
>>>
>>> partner = &lacp->lacpdu.partner;
>>> - if (rte_is_same_ether_addr(&partner->port_params.system,
>>> + if (rte_is_zero_ether_addr(&partner->port_params.system) ||
>>> + rte_is_same_ether_addr(&partner->port_params.system,
>>> &internals->mode4.mac_addr)) {
>>> /* This LACP frame is sending to the bonding port
>>> * so pass it to rx_machine.
>>>
>>
>>
>>
>>
>>
>
>
>
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH] drivers/net/bonding: fix bug for lacp negotiation failed
2020-06-24 10:43 ` Thomas Monjalon
@ 2020-07-10 21:07 ` Thomas Monjalon
2020-07-11 1:42 ` Wei Hu (Xavier)
0 siblings, 1 reply; 5+ messages in thread
From: Thomas Monjalon @ 2020-07-10 21:07 UTC (permalink / raw)
To: luyicai
Cc: dev, chas3, zhangliang, chenchanghu, jerry.lilijun, haifeng.lin,
guohongzhi1, ferruh.yigit, matan, rasland
Still no review.
Please give your real name in the Signed-off so it can be merged.
24/06/2020 12:43, Thomas Monjalon:
> Any review please?
>
> 19/05/2020 13:18, luyicai:
> > When two host is connected directly without any devices like switch,
> > and also enable dedicated tx/rx queues on bonding devices slaves,
> > rx_machine_update would recieving partner lacp negotiation packets,
> > which partner's port mac filled with zeros. So in this situation,
> > it would never go rx_machine branch with correct mac!
> > Thus bond mode 4 will negotiation failed.
> >
> > Signed-off-by: luyicai <luyicai@huawei.com>
> > ---
> > drivers/net/bonding/rte_eth_bond_8023ad.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
> > index b77a37d..2002ec0 100644
> > --- a/drivers/net/bonding/rte_eth_bond_8023ad.c
> > +++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
> > @@ -798,7 +798,8 @@
> > RTE_ASSERT(lacp->lacpdu.subtype == SLOW_SUBTYPE_LACP);
> >
> > partner = &lacp->lacpdu.partner;
> > - if (rte_is_same_ether_addr(&partner->port_params.system,
> > + if (rte_is_zero_ether_addr(&partner->port_params.system) ||
> > + rte_is_same_ether_addr(&partner->port_params.system,
> > &internals->mode4.mac_addr)) {
> > /* This LACP frame is sending to the bonding port
> > * so pass it to rx_machine.
> >
>
>
>
>
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] [PATCH] drivers/net/bonding: fix bug for lacp negotiation failed
2020-05-19 11:18 luyicai
@ 2020-06-24 10:43 ` Thomas Monjalon
2020-07-10 21:07 ` Thomas Monjalon
0 siblings, 1 reply; 5+ messages in thread
From: Thomas Monjalon @ 2020-06-24 10:43 UTC (permalink / raw)
To: dev, chas3
Cc: zhangliang, chenchanghu, jerry.lilijun, haifeng.lin, guohongzhi1,
luyicai, luyicai, ferruh.yigit, matan, rasland
Any review please?
19/05/2020 13:18, luyicai:
> When two host is connected directly without any devices like switch,
> and also enable dedicated tx/rx queues on bonding devices slaves,
> rx_machine_update would recieving partner lacp negotiation packets,
> which partner's port mac filled with zeros. So in this situation,
> it would never go rx_machine branch with correct mac!
> Thus bond mode 4 will negotiation failed.
>
> Signed-off-by: luyicai <luyicai@huawei.com>
> ---
> drivers/net/bonding/rte_eth_bond_8023ad.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
> index b77a37d..2002ec0 100644
> --- a/drivers/net/bonding/rte_eth_bond_8023ad.c
> +++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
> @@ -798,7 +798,8 @@
> RTE_ASSERT(lacp->lacpdu.subtype == SLOW_SUBTYPE_LACP);
>
> partner = &lacp->lacpdu.partner;
> - if (rte_is_same_ether_addr(&partner->port_params.system,
> + if (rte_is_zero_ether_addr(&partner->port_params.system) ||
> + rte_is_same_ether_addr(&partner->port_params.system,
> &internals->mode4.mac_addr)) {
> /* This LACP frame is sending to the bonding port
> * so pass it to rx_machine.
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* [dpdk-dev] [PATCH] drivers/net/bonding: fix bug for lacp negotiation failed
@ 2020-05-19 11:18 luyicai
2020-06-24 10:43 ` Thomas Monjalon
0 siblings, 1 reply; 5+ messages in thread
From: luyicai @ 2020-05-19 11:18 UTC (permalink / raw)
To: dev
Cc: chas3, zhangliang, chenchanghu, jerry.lilijun, haifeng.lin,
guohongzhi1, luyicai
When two host is connected directly without any devices like switch,
and also enable dedicated tx/rx queues on bonding devices slaves,
rx_machine_update would recieving partner lacp negotiation packets,
which partner's port mac filled with zeros. So in this situation,
it would never go rx_machine branch with correct mac!
Thus bond mode 4 will negotiation failed.
Signed-off-by: luyicai <luyicai@huawei.com>
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
index b77a37d..2002ec0 100644
--- a/drivers/net/bonding/rte_eth_bond_8023ad.c
+++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
@@ -798,7 +798,8 @@
RTE_ASSERT(lacp->lacpdu.subtype == SLOW_SUBTYPE_LACP);
partner = &lacp->lacpdu.partner;
- if (rte_is_same_ether_addr(&partner->port_params.system,
+ if (rte_is_zero_ether_addr(&partner->port_params.system) ||
+ rte_is_same_ether_addr(&partner->port_params.system,
&internals->mode4.mac_addr)) {
/* This LACP frame is sending to the bonding port
* so pass it to rx_machine.
--
1.9.5.msysgit.1
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-07-11 1:42 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-19 7:53 [dpdk-dev] [PATCH] drivers/net/bonding: fix bug for lacp negotiation failed luyicai
2020-05-19 11:18 luyicai
2020-06-24 10:43 ` Thomas Monjalon
2020-07-10 21:07 ` Thomas Monjalon
2020-07-11 1:42 ` Wei Hu (Xavier)
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).