Invalid module format when loading ib_* (RHEL 6.4)

Infrastructure & Networking Adapters and Cables InfiniBand/VPI Adapter Cards
1

I installed RHEL 6.4 and upgraded. The current kernel is 2.6.32-358.11.1. Thus I also updated MLNX_OFED_LINUX-2.0-2.0.5-rhel6.4-x86_64.iso with mlnx_add_kernel_support.sh.

My command line is:

/mnt/mlnx_add_kernel_support.sh -m /mnt --make-iso

I installed with mlnxofedinstall after creating new iso image.

However, hca_self_test.ofed issues error at Host Driver Initialization phase.

cat /tmp/hca_self_test_modprobe.output

WARNING: Error inserting ib_core (/lib/modules/2.6.32-358.11.1.el6.x86_64/kernel/drivers/infiniband/core/ib_core.ko): Invalid module format

WARNING: Error inserting ib_mad (/lib/modules/2.6.32-358.11.1.el6.x86_64/kernel/drivers/infiniband/core/ib_mad.ko): Invalid module format

WARNING: Error inserting ib_sa (/lib/modules/2.6.32-358.11.1.el6.x86_64/kernel/drivers/infiniband/core/ib_sa.ko): Invalid module format

WARNING: Error inserting ib_cm (/lib/modules/2.6.32-358.11.1.el6.x86_64/kernel/drivers/infiniband/core/ib_cm.ko): Invalid module format

FATAL: Error inserting ib_ipoib (/lib/modules/2.6.32-358.11.1.el6.x86_64/extra/mlnx-ofa_kernel/drivers/infiniband/ulp/ipoib/ib_ipoib.ko): Invalid module format

What am I wrong and how can I resolve this error?

2

IB software from RedHat is not a solution here. It will work but I didn’t allow me to compile the proper MvAPICH library version.

I know that this might be a litttle bit more work for Mellanox but seems like releasing the new package suited for updated RHEL kernels is the solution.

3

As a first thought, are you sure you need the Mellanox OFED stack?

The yum “Infiniband Support” group included with standard RHEL 6.4 should “just work”:

$ sudo yum groupinstall “Infiniband Support”

Apparently it’s not as optimised for stuff as Mellanox OFED, but if you just need something that works “right out of the box” it’s a good option.

Hopefully that’s useful info.

(note - edited for minor clarification fixes)

4

I have the same problem, however I installed the Lustre kernel (Lustre 2.1, kernel 2.6.32-358.11.1.el6_lustre.x86_64). Did you find the solution?

5

No worres, it was a thought.

And yeah, the “proper” solution here will probably be in form of updated Mellanox OFED stack or similar.