On Mon, May 14, 2018 at 11:11:12AM +0300, Leon Romanovsky wrote:
> +/**
> + * rdma_put_gid_attr - Release reference to the GID attribute
> + * @attr: Pointer to the GID attribute whose reference
> + * needs to be released.
> + *
> + * rdma_put_gid_attr() must be used to release reference whose
> + * reference is acquired using rdma_get_gid_attr() or any APIs
> + * which returns pointer to the ib_gid_attr regardless of link layer
> + * of IB or RoCE.
> + *
> + */
> +void rdma_put_gid_attr(const struct ib_gid_attr *attr)
> +{
> + struct ib_gid_table_entry *entry =
> + container_of(attr, struct ib_gid_table_entry, attr);
> + struct ib_device *device = entry->attr.device;
> + u8 port_num = entry->attr.port_num;
> + struct ib_gid_table *table;
> +
> + if (!rdma_protocol_roce(device, port_num))
> + return;
> + table = device->cache.ports[port_num - rdma_start_port(device)].gid;
> +
> + write_lock_irq(&table->rwlock);
> + kref_put(&entry->kref, schedule_free_roce_gid);
> + write_unlock_irq(&table->rwlock);
Holding a lock here doesn't make any sense to me...
Jason
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
