Map type `BPF_MAP_TYPE_XSKMAP`

This XDP Socket map is a specialized map which references XDP Sockets.

Usage

This map type is used in combination with the bpf_redirect_map helper to redirect traffic to userspace, bypassing the kernel network stack. It is an array style map, where the indices go from 0 to max_entries-1. The values of this map are the file descriptor of specially prepared network sockets.

For details on the usage and preparation of AF_XDP sockets checkout out the concept page and/or kernel docs.

Example

The following examples shows how to use the map and helper from the BPF/kernel side. Important to note is that we always use the rx_queue_index as key, not doing so will cause the packet to be dropped.

struct {
    __uint(type, BPF_MAP_TYPE_XSKMAP);
    __type(key, __u32);
    __type(value, __u32);
    __uint(max_entries, 64);
} xsks_map SEC(".maps");


SEC("xdp")
int xsk_redir_prog(struct xdp_md *ctx)
{
    __u32 index = ctx->rx_queue_index;

    if (bpf_map_lookup_elem(&xsks_map, &index))
            return bpf_redirect_map(&xsks_map, index, 0);
    return XDP_PASS;
}

Attributes

The value_size must always be 4. The key_size must always be 4.

Syscall commands

The following syscall commands work with this map type:

Helper functions

Flags

`BPF_F_NUMA_NODE`

v4.14

When set, the numa_node attribute is respected during map creation.

`BPF_F_RDONLY`

v4.15

Setting this flag will make it so the map can only be read via the syscall interface, but not written to.

For details please check the generic description.

`BPF_F_WRONLY`

v4.15

Setting this flag will make it so the map can only be written to via the syscall interface, but not read from.

Map type BPF_MAP_TYPE_XSKMAP