On Fri, May 27, 2022 at 10:28 AM Romain Delhomel <romain.delhomel@6wind.com>
wrote:

> At device probe, the fslmc bus driver calls rte_vfio_get_group_fd() to
> get a fd associated to a vfio group. This function first checks if the
> group is already opened, else it opens /dev/vfio/%u, and increases the
> number of active groups in default_vfio_cfg (which references the
> default vfio container).
>
> When adding the first group to a vfio_cfg, the caller is supposed to
> pick an IOMMU type and set up DMA mappings for container, as it's done
> by pci bus, but it is not done here. Instead, a new container is created
> and used.
>
> This prevents the pci bus driver, which use the default_vfio_cfg
> container, to configure the container because
> default_vfio_cfg->active_group > 1.
>
> This patch fixes the issue by always creating a new container (and its
> associated vfio_cfg) and bind the group to it.
>
> Fixes: a69f79300262 ("bus/fslmc: support multi VFIO group")
> Signed-off-by: Romain Delhomel <romain.delhomel@6wind.com>
> ---
>  drivers/bus/fslmc/fslmc_vfio.c | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/bus/fslmc/fslmc_vfio.c
> b/drivers/bus/fslmc/fslmc_vfio.c
> index b4704eeae4e2..abe1cab2ee20 100644
> --- a/drivers/bus/fslmc/fslmc_vfio.c
> +++ b/drivers/bus/fslmc/fslmc_vfio.c
> @@ -979,6 +979,7 @@ fslmc_vfio_setup_group(void)
>  {
>   int groupid;
>   int ret;
> + int vfio_container_fd;
>   struct vfio_group_status status = { .argsz = sizeof(status) };
>
>   /* if already done once */
> @@ -997,8 +998,15 @@ fslmc_vfio_setup_group(void)
>   return 0;
>   }
>
> + ret = rte_vfio_container_create();
> + if (ret < 0) {
> + DPAA2_BUS_ERR("Failed to open VFIO container");
> + return ret;
> + }
> + vfio_container_fd = ret;
> +
>   /* Get the actual group fd */
> - ret = rte_vfio_get_group_fd(groupid);
> + ret = rte_vfio_container_group_bind(vfio_container_fd, groupid);
>   if (ret < 0)
>   return ret;
>   vfio_group.fd = ret;
> --
> 2.30.2
>