* i40e XDP program stops transmitting after link down/up
@ 2024-01-30 14:17 Seth Forshee
2024-01-30 16:14 ` [Intel-wired-lan] " Paul Menzel
0 siblings, 1 reply; 7+ messages in thread
From: Seth Forshee @ 2024-01-30 14:17 UTC (permalink / raw)
To: Jesse Brandeburg, Tony Nguyen; +Cc: intel-wired-lan, netdev
I got a inquiry from a colleague about a behavior he's seeing with i40e
but not with other NICs. The interfaces are bonded with a XDP
load-balancer program attached to them. After 'ip link set ethX down; ip
link set ethX up' on one of the interfaces the XDP program on that
interface is no longer transmitting packets. He found that tx starts
again after running 'sudo ethtool -t ethX'.
There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
timeout' message in dmesg when disabling the interface. I've included
the relevant portions from dmesg below.
This was first observed with a 6.1 kernel, but we've confirmed that the
behavior is the same in 6.7. I realize the firmware is pretty old, so
far our attempts to update the NVM have failed.
Thanks,
Seth
[ 0.000000] Linux version 6.7.0 (root@616a530b3729) (gcc (Ubuntu 11.3.0-1ubuntu1~22.04) 11.3.0, GNU ld (GNU Binutils for Ubuntu) 2.38) #2 SMP Thu Jan 25 10:37:21 EST 2024
...
[ 9.038171] i40e: Intel(R) Ethernet Connection XL710 Network Driver
[ 9.044447] i40e: Copyright (c) 2013 - 2019 Intel Corporation.
...
[ 9.064833] i40e 0000:d8:00.0: fw 7.0.50775 api 1.8 nvm 7.00 0x80004c97 1.2154.0 [8086:1583] [8086:0002]
...
[ 9.320886] i40e 0000:d8:00.0: MAC address: xx:xx:xx:xx:xx:xx
[ 9.327030] i40e 0000:d8:00.0: FW LLDP is enabled
[ 9.344331] i40e 0000:d8:00.0 eth0: NIC Link is Up, 40 Gbps Full Duplex, Flow Control: None
[ 9.355552] i40e 0000:d8:00.0: PCI-Express: Speed 8.0GT/s Width x8
[ 9.374074] i40e 0000:d8:00.0: Features: PF-id[0] VFs: 64 VSIs: 66 QP: 32 RSS FD_ATR FD_SB NTUPLE DCB VxLAN Geneve PTP VEPA
...
[ 9.401522] i40e 0000:d8:00.1: fw 7.0.50775 api 1.8 nvm 7.00 0x80004c97 1.2154.0 [8086:1583] [8086:0002]
...
[ 9.652066] i40e 0000:d8:00.1: MAC address: xx:xx:xx:xx:xx:xx
[ 9.658040] i40e 0000:d8:00.1: FW LLDP is enabled
[ 9.688622] i40e 0000:d8:00.1 eth1: NIC Link is Up, 40 Gbps Full Duplex, Flow Control: None
[ 9.699822] i40e 0000:d8:00.1: PCI-Express: Speed 8.0GT/s Width x8
[ 9.719259] i40e 0000:d8:00.1: Features: PF-id[1] VFs: 64 VSIs: 66 QP: 32 RSS FD_ATR FD_SB NTUPLE DCB VxLAN Geneve PTP VEPA
[ 9.401522] i40e 0000:d8:00.1: fw 7.0.50775 api 1.8 nvm 7.00 0x80004c97 1.2154.0 [8086:1583] [8086:0002]
...
[ 13.948968] i40e 0000:d8:00.0: FW LLDP is disabled, attempting SW DCB
[ 13.956753] i40e 0000:d8:00.0: SW DCB initialization succeeded.
[ 13.970005] i40e 0000:d8:00.0: FW LLDP is disabled
...
[ 14.113004] i40e 0000:d8:00.1: FW LLDP is disabled, attempting SW DCB
[ 14.120800] i40e 0000:d8:00.1: SW DCB initialization succeeded.
[ 14.133940] i40e 0000:d8:00.1: FW LLDP is disabled
...
[ 14.267874] bonding: bond0 is being created...
[ 14.979070] i40e 0000:d8:00.1 eth1: set new mac address xx:xx:xx:xx:xx:xx
[ 14.994445] bond0: (slave eth1): Enslaving as a backup interface with an up link
[ 14.994681] i40e 0000:d8:00.0 eth0: set new mac address xx:xx:xx:xx:xx:xx
[ 15.010359] bond0: (slave eth0): Enslaving as a backup interface with an up link
[ 15.912874] i40e 0000:d8:00.0: Stop LLDP AQ command failed =0x1
[ 15.944842] i40e 0000:d8:00.1: Stop LLDP AQ command failed =0x1
[ 41.262871] 8021q: 802.1Q VLAN Support v1.8
[ 41.262890] 8021q: adding VLAN 0 to HW filter on device eth0
[ 41.262902] 8021q: adding VLAN 0 to HW filter on device eth1
[ 41.262914] 8021q: adding VLAN 0 to HW filter on device bond0
[ 48.272456] i40e 0000:d8:00.0: FW LLDP is disabled, attempting SW DCB
[ 48.280233] i40e 0000:d8:00.0: SW DCB initialization succeeded.
[ 48.307415] i40e 0000:d8:00.0: User requested queue count/HW max RSS count: 12/32
[ 48.440266] i40e 0000:d8:00.1: FW LLDP is disabled, attempting SW DCB
[ 48.448025] i40e 0000:d8:00.1: SW DCB initialization succeeded.
[ 48.475051] i40e 0000:d8:00.1: User requested queue count/HW max RSS count: 12/32
[ 58.935900] i40e 0000:d8:00.0: FW LLDP is disabled, attempting SW DCB
[ 58.945123] i40e 0000:d8:00.0: SW DCB initialization succeeded.
[ 59.131772] i40e 0000:d8:00.1: FW LLDP is disabled, attempting SW DCB
[ 59.139560] i40e 0000:d8:00.1: SW DCB initialization succeeded.
[ 336.363825] i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable timeout
[ 336.603619] bond0: (slave eth1): link status definitely down, disabling slave
[ 345.464976] 8021q: adding VLAN 0 to HW filter on device eth1
[ 345.547358] bond0: (slave eth1): link status definitely up, 40000 Mbps full duplex
[ 345.547379] bond0: active interface up!
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Intel-wired-lan] i40e XDP program stops transmitting after link down/up
2024-01-30 14:17 i40e XDP program stops transmitting after link down/up Seth Forshee
@ 2024-01-30 16:14 ` Paul Menzel
2024-01-30 16:59 ` Seth Forshee
0 siblings, 1 reply; 7+ messages in thread
From: Paul Menzel @ 2024-01-30 16:14 UTC (permalink / raw)
To: Seth Forshee; +Cc: Jesse Brandeburg, Tony Nguyen, netdev, intel-wired-lan
Dear Seth,
Thank you for bring this up.
Am 30.01.24 um 15:17 schrieb Seth Forshee:
> I got a inquiry from a colleague about a behavior he's seeing with i40e
> but not with other NICs. The interfaces are bonded with a XDP
> load-balancer program attached to them. After 'ip link set ethX down; ip
> link set ethX up' on one of the interfaces the XDP program on that
> interface is no longer transmitting packets. He found that tx starts
> again after running 'sudo ethtool -t ethX'.
>
> There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
> timeout' message in dmesg when disabling the interface. I've included
> the relevant portions from dmesg below.
>
> This was first observed with a 6.1 kernel, but we've confirmed that the
> behavior is the same in 6.7. I realize the firmware is pretty old, so
> far our attempts to update the NVM have failed.
Does that mean, the problem didn’t happen before Linux 6.1? If so, if
you have the reproducer and the time, bisecting the issue is normally
the fastest way to solve the issue.
Kind regards,
Paul
> [ 0.000000] Linux version 6.7.0 (root@616a530b3729) (gcc (Ubuntu 11.3.0-1ubuntu1~22.04) 11.3.0, GNU ld (GNU Binutils for Ubuntu) 2.38) #2 SMP Thu Jan 25 10:37:21 EST 2024
> ...
> [ 9.038171] i40e: Intel(R) Ethernet Connection XL710 Network Driver
> [ 9.044447] i40e: Copyright (c) 2013 - 2019 Intel Corporation.
> ...
> [ 9.064833] i40e 0000:d8:00.0: fw 7.0.50775 api 1.8 nvm 7.00 0x80004c97 1.2154.0 [8086:1583] [8086:0002]
> ...
> [ 9.320886] i40e 0000:d8:00.0: MAC address: xx:xx:xx:xx:xx:xx
> [ 9.327030] i40e 0000:d8:00.0: FW LLDP is enabled
> [ 9.344331] i40e 0000:d8:00.0 eth0: NIC Link is Up, 40 Gbps Full Duplex, Flow Control: None
> [ 9.355552] i40e 0000:d8:00.0: PCI-Express: Speed 8.0GT/s Width x8
> [ 9.374074] i40e 0000:d8:00.0: Features: PF-id[0] VFs: 64 VSIs: 66 QP: 32 RSS FD_ATR FD_SB NTUPLE DCB VxLAN Geneve PTP VEPA
> ...
> [ 9.401522] i40e 0000:d8:00.1: fw 7.0.50775 api 1.8 nvm 7.00 0x80004c97 1.2154.0 [8086:1583] [8086:0002]
> ...
> [ 9.652066] i40e 0000:d8:00.1: MAC address: xx:xx:xx:xx:xx:xx
> [ 9.658040] i40e 0000:d8:00.1: FW LLDP is enabled
> [ 9.688622] i40e 0000:d8:00.1 eth1: NIC Link is Up, 40 Gbps Full Duplex, Flow Control: None
> [ 9.699822] i40e 0000:d8:00.1: PCI-Express: Speed 8.0GT/s Width x8
> [ 9.719259] i40e 0000:d8:00.1: Features: PF-id[1] VFs: 64 VSIs: 66 QP: 32 RSS FD_ATR FD_SB NTUPLE DCB VxLAN Geneve PTP VEPA
> [ 9.401522] i40e 0000:d8:00.1: fw 7.0.50775 api 1.8 nvm 7.00 0x80004c97 1.2154.0 [8086:1583] [8086:0002]
> ...
> [ 13.948968] i40e 0000:d8:00.0: FW LLDP is disabled, attempting SW DCB
> [ 13.956753] i40e 0000:d8:00.0: SW DCB initialization succeeded.
> [ 13.970005] i40e 0000:d8:00.0: FW LLDP is disabled
> ...
> [ 14.113004] i40e 0000:d8:00.1: FW LLDP is disabled, attempting SW DCB
> [ 14.120800] i40e 0000:d8:00.1: SW DCB initialization succeeded.
> [ 14.133940] i40e 0000:d8:00.1: FW LLDP is disabled
> ...
> [ 14.267874] bonding: bond0 is being created...
> [ 14.979070] i40e 0000:d8:00.1 eth1: set new mac address xx:xx:xx:xx:xx:xx
> [ 14.994445] bond0: (slave eth1): Enslaving as a backup interface with an up link
> [ 14.994681] i40e 0000:d8:00.0 eth0: set new mac address xx:xx:xx:xx:xx:xx
> [ 15.010359] bond0: (slave eth0): Enslaving as a backup interface with an up link
> [ 15.912874] i40e 0000:d8:00.0: Stop LLDP AQ command failed =0x1
> [ 15.944842] i40e 0000:d8:00.1: Stop LLDP AQ command failed =0x1
> [ 41.262871] 8021q: 802.1Q VLAN Support v1.8
> [ 41.262890] 8021q: adding VLAN 0 to HW filter on device eth0
> [ 41.262902] 8021q: adding VLAN 0 to HW filter on device eth1
> [ 41.262914] 8021q: adding VLAN 0 to HW filter on device bond0
> [ 48.272456] i40e 0000:d8:00.0: FW LLDP is disabled, attempting SW DCB
> [ 48.280233] i40e 0000:d8:00.0: SW DCB initialization succeeded.
> [ 48.307415] i40e 0000:d8:00.0: User requested queue count/HW max RSS count: 12/32
> [ 48.440266] i40e 0000:d8:00.1: FW LLDP is disabled, attempting SW DCB
> [ 48.448025] i40e 0000:d8:00.1: SW DCB initialization succeeded.
> [ 48.475051] i40e 0000:d8:00.1: User requested queue count/HW max RSS count: 12/32
> [ 58.935900] i40e 0000:d8:00.0: FW LLDP is disabled, attempting SW DCB
> [ 58.945123] i40e 0000:d8:00.0: SW DCB initialization succeeded.
> [ 59.131772] i40e 0000:d8:00.1: FW LLDP is disabled, attempting SW DCB
> [ 59.139560] i40e 0000:d8:00.1: SW DCB initialization succeeded.
> [ 336.363825] i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable timeout
> [ 336.603619] bond0: (slave eth1): link status definitely down, disabling slave
> [ 345.464976] 8021q: adding VLAN 0 to HW filter on device eth1
> [ 345.547358] bond0: (slave eth1): link status definitely up, 40000 Mbps full duplex
> [ 345.547379] bond0: active interface up!
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Intel-wired-lan] i40e XDP program stops transmitting after link down/up
2024-01-30 16:14 ` [Intel-wired-lan] " Paul Menzel
@ 2024-01-30 16:59 ` Seth Forshee
2024-01-30 19:28 ` Maciej Fijalkowski
0 siblings, 1 reply; 7+ messages in thread
From: Seth Forshee @ 2024-01-30 16:59 UTC (permalink / raw)
To: Paul Menzel; +Cc: Jesse Brandeburg, Tony Nguyen, netdev, intel-wired-lan
On Tue, Jan 30, 2024 at 05:14:23PM +0100, Paul Menzel wrote:
> Dear Seth,
>
>
> Thank you for bring this up.
>
> Am 30.01.24 um 15:17 schrieb Seth Forshee:
> > I got a inquiry from a colleague about a behavior he's seeing with i40e
> > but not with other NICs. The interfaces are bonded with a XDP
> > load-balancer program attached to them. After 'ip link set ethX down; ip
> > link set ethX up' on one of the interfaces the XDP program on that
> > interface is no longer transmitting packets. He found that tx starts
> > again after running 'sudo ethtool -t ethX'.
> >
> > There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
> > timeout' message in dmesg when disabling the interface. I've included
> > the relevant portions from dmesg below.
> >
> > This was first observed with a 6.1 kernel, but we've confirmed that the
> > behavior is the same in 6.7. I realize the firmware is pretty old, so
> > far our attempts to update the NVM have failed.
>
> Does that mean, the problem didn’t happen before Linux 6.1? If so, if you
> have the reproducer and the time, bisecting the issue is normally the
> fastest way to solve the issue.
No, sorry, I should have worded that better. I meant that they were
using 6.1 when they noticed the issue, not that kernels before 6.1 did
not have that issue. We've also tried a 5.15 kernel build now and still
see the issue there, we haven't tested anything older than that.
Thanks,
Seth
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Intel-wired-lan] i40e XDP program stops transmitting after link down/up
2024-01-30 16:59 ` Seth Forshee
@ 2024-01-30 19:28 ` Maciej Fijalkowski
2024-01-30 22:06 ` Maciej Fijalkowski
0 siblings, 1 reply; 7+ messages in thread
From: Maciej Fijalkowski @ 2024-01-30 19:28 UTC (permalink / raw)
To: Seth Forshee
Cc: Paul Menzel, Jesse Brandeburg, Tony Nguyen, netdev,
intel-wired-lan
On Tue, Jan 30, 2024 at 10:59:13AM -0600, Seth Forshee wrote:
> On Tue, Jan 30, 2024 at 05:14:23PM +0100, Paul Menzel wrote:
> > Dear Seth,
> >
> >
> > Thank you for bring this up.
> >
> > Am 30.01.24 um 15:17 schrieb Seth Forshee:
> > > I got a inquiry from a colleague about a behavior he's seeing with i40e
> > > but not with other NICs. The interfaces are bonded with a XDP
> > > load-balancer program attached to them. After 'ip link set ethX down; ip
> > > link set ethX up' on one of the interfaces the XDP program on that
> > > interface is no longer transmitting packets. He found that tx starts
> > > again after running 'sudo ethtool -t ethX'.
> > >
> > > There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
> > > timeout' message in dmesg when disabling the interface. I've included
> > > the relevant portions from dmesg below.
> > >
> > > This was first observed with a 6.1 kernel, but we've confirmed that the
> > > behavior is the same in 6.7. I realize the firmware is pretty old, so
> > > far our attempts to update the NVM have failed.
> >
> > Does that mean, the problem didn’t happen before Linux 6.1? If so, if you
> > have the reproducer and the time, bisecting the issue is normally the
> > fastest way to solve the issue.
>
> No, sorry, I should have worded that better. I meant that they were
> using 6.1 when they noticed the issue, not that kernels before 6.1 did
> not have that issue. We've also tried a 5.15 kernel build now and still
> see the issue there, we haven't tested anything older than that.
Hey Seth,
I am observing same thing on my side with xdpsock in txonly mode, so I'll
take a look at this and will keep you updated.
>
> Thanks,
> Seth
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Intel-wired-lan] i40e XDP program stops transmitting after link down/up
2024-01-30 19:28 ` Maciej Fijalkowski
@ 2024-01-30 22:06 ` Maciej Fijalkowski
2024-01-31 16:26 ` Seth Forshee
0 siblings, 1 reply; 7+ messages in thread
From: Maciej Fijalkowski @ 2024-01-30 22:06 UTC (permalink / raw)
To: Seth Forshee
Cc: Paul Menzel, Jesse Brandeburg, Tony Nguyen, netdev,
intel-wired-lan
On Tue, Jan 30, 2024 at 08:28:24PM +0100, Maciej Fijalkowski wrote:
> On Tue, Jan 30, 2024 at 10:59:13AM -0600, Seth Forshee wrote:
> > On Tue, Jan 30, 2024 at 05:14:23PM +0100, Paul Menzel wrote:
> > > Dear Seth,
> > >
> > >
> > > Thank you for bring this up.
> > >
> > > Am 30.01.24 um 15:17 schrieb Seth Forshee:
> > > > I got a inquiry from a colleague about a behavior he's seeing with i40e
> > > > but not with other NICs. The interfaces are bonded with a XDP
> > > > load-balancer program attached to them. After 'ip link set ethX down; ip
> > > > link set ethX up' on one of the interfaces the XDP program on that
> > > > interface is no longer transmitting packets. He found that tx starts
> > > > again after running 'sudo ethtool -t ethX'.
> > > >
> > > > There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
> > > > timeout' message in dmesg when disabling the interface. I've included
> > > > the relevant portions from dmesg below.
> > > >
> > > > This was first observed with a 6.1 kernel, but we've confirmed that the
> > > > behavior is the same in 6.7. I realize the firmware is pretty old, so
> > > > far our attempts to update the NVM have failed.
> > >
> > > Does that mean, the problem didn’t happen before Linux 6.1? If so, if you
> > > have the reproducer and the time, bisecting the issue is normally the
> > > fastest way to solve the issue.
> >
> > No, sorry, I should have worded that better. I meant that they were
> > using 6.1 when they noticed the issue, not that kernels before 6.1 did
> > not have that issue. We've also tried a 5.15 kernel build now and still
> > see the issue there, we haven't tested anything older than that.
>
> Hey Seth,
>
> I am observing same thing on my side with xdpsock in txonly mode, so I'll
> take a look at this and will keep you updated.
Can you try this diff on your side?
From 1d3b2d92506c0af4bea843c046bdef34f3f66893 Mon Sep 17 00:00:00 2001
From: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Date: Tue, 30 Jan 2024 23:01:28 +0100
Subject: [PATCH net] i40e: stop XDP txqs
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 6e7fd473abfd..08ef88783762 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -4926,13 +4926,16 @@ int i40e_vsi_start_rings(struct i40e_vsi *vsi)
void i40e_vsi_stop_rings(struct i40e_vsi *vsi)
{
struct i40e_pf *pf = vsi->back;
+ u32 alloc_queue_pairs;
int pf_q, err, q_end;
/* When port TX is suspended, don't wait */
if (test_bit(__I40E_PORT_SUSPENDED, vsi->back->state))
return i40e_vsi_stop_rings_no_wait(vsi);
- q_end = vsi->base_queue + vsi->num_queue_pairs;
+ alloc_queue_pairs = vsi->alloc_queue_pairs *
+ (i40e_enabled_xdp_vsi(vsi) ? 2 : 1);
+ q_end = vsi->base_queue + alloc_queue_pairs;
for (pf_q = vsi->base_queue; pf_q < q_end; pf_q++)
i40e_pre_tx_queue_cfg(&pf->hw, (u32)pf_q, false);
--
2.34.1
>
> >
> > Thanks,
> > Seth
> >
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [Intel-wired-lan] i40e XDP program stops transmitting after link down/up
2024-01-30 22:06 ` Maciej Fijalkowski
@ 2024-01-31 16:26 ` Seth Forshee
2024-01-31 16:38 ` Maciej Fijalkowski
0 siblings, 1 reply; 7+ messages in thread
From: Seth Forshee @ 2024-01-31 16:26 UTC (permalink / raw)
To: Maciej Fijalkowski
Cc: Paul Menzel, Jesse Brandeburg, Tony Nguyen, netdev,
intel-wired-lan
On Tue, Jan 30, 2024 at 11:06:07PM +0100, Maciej Fijalkowski wrote:
> On Tue, Jan 30, 2024 at 08:28:24PM +0100, Maciej Fijalkowski wrote:
> > On Tue, Jan 30, 2024 at 10:59:13AM -0600, Seth Forshee wrote:
> > > On Tue, Jan 30, 2024 at 05:14:23PM +0100, Paul Menzel wrote:
> > > > Dear Seth,
> > > >
> > > >
> > > > Thank you for bring this up.
> > > >
> > > > Am 30.01.24 um 15:17 schrieb Seth Forshee:
> > > > > I got a inquiry from a colleague about a behavior he's seeing with i40e
> > > > > but not with other NICs. The interfaces are bonded with a XDP
> > > > > load-balancer program attached to them. After 'ip link set ethX down; ip
> > > > > link set ethX up' on one of the interfaces the XDP program on that
> > > > > interface is no longer transmitting packets. He found that tx starts
> > > > > again after running 'sudo ethtool -t ethX'.
> > > > >
> > > > > There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
> > > > > timeout' message in dmesg when disabling the interface. I've included
> > > > > the relevant portions from dmesg below.
> > > > >
> > > > > This was first observed with a 6.1 kernel, but we've confirmed that the
> > > > > behavior is the same in 6.7. I realize the firmware is pretty old, so
> > > > > far our attempts to update the NVM have failed.
> > > >
> > > > Does that mean, the problem didn’t happen before Linux 6.1? If so, if you
> > > > have the reproducer and the time, bisecting the issue is normally the
> > > > fastest way to solve the issue.
> > >
> > > No, sorry, I should have worded that better. I meant that they were
> > > using 6.1 when they noticed the issue, not that kernels before 6.1 did
> > > not have that issue. We've also tried a 5.15 kernel build now and still
> > > see the issue there, we haven't tested anything older than that.
> >
> > Hey Seth,
> >
> > I am observing same thing on my side with xdpsock in txonly mode, so I'll
> > take a look at this and will keep you updated.
>
> Can you try this diff on your side?
That fixes the problem!
Thanks,
Seth
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Intel-wired-lan] i40e XDP program stops transmitting after link down/up
2024-01-31 16:26 ` Seth Forshee
@ 2024-01-31 16:38 ` Maciej Fijalkowski
0 siblings, 0 replies; 7+ messages in thread
From: Maciej Fijalkowski @ 2024-01-31 16:38 UTC (permalink / raw)
To: Seth Forshee
Cc: Paul Menzel, Jesse Brandeburg, Tony Nguyen, netdev,
intel-wired-lan, magnus.karlsson
On Wed, Jan 31, 2024 at 10:26:07AM -0600, Seth Forshee wrote:
> On Tue, Jan 30, 2024 at 11:06:07PM +0100, Maciej Fijalkowski wrote:
> > On Tue, Jan 30, 2024 at 08:28:24PM +0100, Maciej Fijalkowski wrote:
> > > On Tue, Jan 30, 2024 at 10:59:13AM -0600, Seth Forshee wrote:
> > > > On Tue, Jan 30, 2024 at 05:14:23PM +0100, Paul Menzel wrote:
> > > > > Dear Seth,
> > > > >
> > > > >
> > > > > Thank you for bring this up.
> > > > >
> > > > > Am 30.01.24 um 15:17 schrieb Seth Forshee:
> > > > > > I got a inquiry from a colleague about a behavior he's seeing with i40e
> > > > > > but not with other NICs. The interfaces are bonded with a XDP
> > > > > > load-balancer program attached to them. After 'ip link set ethX down; ip
> > > > > > link set ethX up' on one of the interfaces the XDP program on that
> > > > > > interface is no longer transmitting packets. He found that tx starts
> > > > > > again after running 'sudo ethtool -t ethX'.
> > > > > >
> > > > > > There's a 'i40e 0000:d8:00.1: VSI seid 391 XDP Tx ring 0 disable
> > > > > > timeout' message in dmesg when disabling the interface. I've included
> > > > > > the relevant portions from dmesg below.
> > > > > >
> > > > > > This was first observed with a 6.1 kernel, but we've confirmed that the
> > > > > > behavior is the same in 6.7. I realize the firmware is pretty old, so
> > > > > > far our attempts to update the NVM have failed.
> > > > >
> > > > > Does that mean, the problem didn’t happen before Linux 6.1? If so, if you
> > > > > have the reproducer and the time, bisecting the issue is normally the
> > > > > fastest way to solve the issue.
> > > >
> > > > No, sorry, I should have worded that better. I meant that they were
> > > > using 6.1 when they noticed the issue, not that kernels before 6.1 did
> > > > not have that issue. We've also tried a 5.15 kernel build now and still
> > > > see the issue there, we haven't tested anything older than that.
> > >
> > > Hey Seth,
> > >
> > > I am observing same thing on my side with xdpsock in txonly mode, so I'll
> > > take a look at this and will keep you updated.
> >
> > Can you try this diff on your side?
>
> That fixes the problem!
Awesome. I'll send a proper fix then and will keep you in loop.
>
> Thanks,
> Seth
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2024-01-31 16:39 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-01-30 14:17 i40e XDP program stops transmitting after link down/up Seth Forshee
2024-01-30 16:14 ` [Intel-wired-lan] " Paul Menzel
2024-01-30 16:59 ` Seth Forshee
2024-01-30 19:28 ` Maciej Fijalkowski
2024-01-30 22:06 ` Maciej Fijalkowski
2024-01-31 16:26 ` Seth Forshee
2024-01-31 16:38 ` Maciej Fijalkowski
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).