Re: [dpdk-users] common_mlx5: Failed to load driver = net_mlx5 Error

[Raslan Darawsheh <rasland@nvidia.com>]

Hi Pravein,

Do you happen to have any errors in dmesg output ?
Also, can you post the output of ibdev2netdev ?

Kindest regards,
Raslan Darawsheh

From: Pravein GK <praveingk@gmail.com>
Sent: Thursday, January 28, 2021 10:36 AM
To: Raslan Darawsheh <rasland@nvidia.com>
Cc: users@dpdk.org
Subject: Re: [dpdk-users] common_mlx5: Failed to load driver = net_mlx5 Error

Hi Raslan,

Thanks for your reply.

The lsmod o/p is below :

$ lsmod | grep mlx5_
mlx5_ib               364544  0
ib_uverbs             131072  2 rdma_ucm,mlx5_ib
mlx5_core            1294336  1 mlx5_ib
mlxfw                  24576  1 mlx5_core
psample                20480  1 mlx5_core
devlink                45056  1 mlx5_core
ib_core               311296  8 rdma_cm,ib_ipoib,iw_cm,ib_umad,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm
mlx_compat             45056  10 rdma_cm,ib_ipoib,iw_cm,ib_umad,ib_core,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm,mlx5_core
ptp                    20480  1 mlx5_core

I had tried the force restart, and even after that , I get the same error.

Thanks,
Pravein



On Thu, Jan 28, 2021 at 1:52 PM Raslan Darawsheh <rasland@nvidia.com<mailto:rasland@nvidia.com>> wrote:
Hi Parvein,

Can you kindly confirm that the kernel drivers (mlx5_ib, mlx5_core) are loaded?
(lsmod | grep mlx5_)

If not please try to restart the kernel driver:

/etc/init.d/openibd force-restart

Kindest regards,
Raslan Darawsheh

> -----Original Message-----
> From: users <users-bounces@dpdk.org<mailto:users-bounces@dpdk.org>> On Behalf Of Pravein GK
> Sent: Thursday, January 28, 2021 7:49 AM
> To: users@dpdk.org<mailto:users@dpdk.org>
> Subject: [dpdk-users] common_mlx5: Failed to load driver = net_mlx5 Error
>
> Hello All,
>
> I am using DPDK 20.08 with Mellanox Connect X 5 NICs.
> I have installed the latest OFED drivers with "--dpdk --upstream-libs".
> However, while launching a dpdk application, I get the below error :
>
> EAL: Detected 24 lcore(s)
> > EAL: Detected 2 NUMA nodes
> > EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
> > EAL: Selected IOVA mode 'PA'
> > EAL: No available hugepages reported in hugepages-1048576kB
> > EAL: Probing VFIO support...
> > EAL:   Invalid NUMA socket, default to 0
> >
> >
> > *EAL: Probe PCI driver: mlx5_pci (15b3:1019) device: 0000:1f:00.0 (socket
> > 0)net_mlx5: no Verbs device matches PCI device 0000:1f:00.0, are kernel
> > drivers loaded?common_mlx5: Failed to load driver = net_mlx5.*
> > EAL: Requested device 0000:1f:00.0 cannot be used
> > EAL:   Invalid NUMA socket, default to 0
> >
> >
> > *EAL: Probe PCI driver: mlx5_pci (15b3:1019) device: 0000:1f:00.1 (socket
> > 0)net_mlx5: no Verbs device matches PCI device 0000:1f:00.1, are kernel
> > drivers loaded?common_mlx5: Failed to load driver = net_mlx5.*
> > EAL: Requested device 0000:1f:00.1 cannot be used
> > EAL: No legacy callbacks, legacy socket not created
> > EAL: Error - exiting with code: 1
> >   Cause: Cannot create mbuf pool , error=Invalid argument
>
>
> I am stuck with this error, and unable to proceed for a long time despite
> trying most of the things suggested in forums. Please help.
>
> Thanks,
> Pravein