On Thu, Mar 27, 2025 at 8:57 AM Yang Ming <ming.1.yang@nokia-sbell.com> wrote:

DPDK detect vfio container according the existence of vfio
module. But for container with non-privileged mode, there is
possibility that no VFIO_DIR(/dev/vfio) mapping from host to
container when host have both Intel NIC and Mellanox NIC but
this conntainer only allocate VFs from Mellanox NIC.
In this case, vfio kernel module has already been loaded from
the host.
This scenario will cause the error log occurs in DPDK primary
process as below:
'EAL:   cannot open VFIO container, error 2 (No such file or
directory)'
'EAL: VFIO support could not be initialized'
Because `rte_vfio_enable()` call `rte_vfio_get_container_fd()`
to execute `vfio_container_fd = open(VFIO_CONTAINER_PATH,
O_RDWR);` but VFIO_CONTAINER_PATH(/dev/vfio/vfio) doesn't exist
in this container.
This scenario will also lead to the delay of DPDK secondary
process because `default_vfio_cfg->vfio_enabled = 0` and
`default_vfio_cfg->vfio_container_fd = -1`, socket error will
be set in DPDK primary process when it sync this info to
the secondary process.
This patch use to skip this kind of useless detection for this
scenario.

Signed-off-by: Yang Ming <ming.1.yang@nokia-sbell.com>

With such a change, is the check on the passed kernel module still needed?

Hi David,

Thanks for your comment.

Yes. We still need this checks:
1. Module check (rte_eal_check_module(modname)) ensures the host has the VFIO driver loaded.
2. Directory check (opendir("/dev/vfio")) skips the open call in unprivileged containers without /dev/vfio, avoiding a noisy error.
This patch adds the second check. Please let me know if you'd like any more details!

Best regards,
Yang Ming