passt: New design and implementation with native Layer 4 sockets
This is a reimplementation, partially building on the earlier draft,
that uses L4 sockets (SOCK_DGRAM, SOCK_STREAM) instead of SOCK_RAW,
providing L4-L2 translation functionality without requiring any
security capability.
Conceptually, this follows the design presented at:
https://gitlab.com/abologna/kubevirt-and-kvm/-/blob/master/Networking.md
The most significant novelty here comes from TCP and UDP translation
layers. In particular, the TCP state and translation logic follows
the intent of being minimalistic, without reimplementing a full TCP
stack in either direction, and synchronising as much as possible the
TCP dynamic and flows between guest and host kernel.
Another important introduction concerns addressing, port translation
and forwarding. The Layer 4 implementations now attempt to bind on
all unbound ports, in order to forward connections in a transparent
way.
While at it:
- the qemu 'tap' back-end can't be used as-is by qrap anymore,
because of explicit checks now introduced in qemu to ensure that
the corresponding file descriptor is actually a tap device. For
this reason, qrap now operates on a 'socket' back-end type,
accounting for and building the additional header reporting
frame length
- provide a demo script that sets up namespaces, addresses and
routes, and starts the daemon. A virtual machine started in the
network namespace, wrapped by qrap, will now directly interface
with passt and communicate using Layer 4 sockets provided by the
host kernel.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
2021-02-16 07:25:09 +01:00
|
|
|
// SPDX-License-Identifier: AGPL-3.0-or-later
|
|
|
|
|
2020-07-21 10:48:24 +02:00
|
|
|
/* PASST - Plug A Simple Socket Transport
|
|
|
|
*
|
|
|
|
* ndp.c - NDP support for PASST
|
|
|
|
*
|
passt: New design and implementation with native Layer 4 sockets
This is a reimplementation, partially building on the earlier draft,
that uses L4 sockets (SOCK_DGRAM, SOCK_STREAM) instead of SOCK_RAW,
providing L4-L2 translation functionality without requiring any
security capability.
Conceptually, this follows the design presented at:
https://gitlab.com/abologna/kubevirt-and-kvm/-/blob/master/Networking.md
The most significant novelty here comes from TCP and UDP translation
layers. In particular, the TCP state and translation logic follows
the intent of being minimalistic, without reimplementing a full TCP
stack in either direction, and synchronising as much as possible the
TCP dynamic and flows between guest and host kernel.
Another important introduction concerns addressing, port translation
and forwarding. The Layer 4 implementations now attempt to bind on
all unbound ports, in order to forward connections in a transparent
way.
While at it:
- the qemu 'tap' back-end can't be used as-is by qrap anymore,
because of explicit checks now introduced in qemu to ensure that
the corresponding file descriptor is actually a tap device. For
this reason, qrap now operates on a 'socket' back-end type,
accounting for and building the additional header reporting
frame length
- provide a demo script that sets up namespaces, addresses and
routes, and starts the daemon. A virtual machine started in the
network namespace, wrapped by qrap, will now directly interface
with passt and communicate using Layer 4 sockets provided by the
host kernel.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
2021-02-16 07:25:09 +01:00
|
|
|
* Copyright (c) 2020-2021 Red Hat GmbH
|
2020-07-21 10:48:24 +02:00
|
|
|
* Author: Stefano Brivio <sbrivio@redhat.com>
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <stddef.h>
|
|
|
|
#include <stdint.h>
|
|
|
|
#include <unistd.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <linux/if_ether.h>
|
|
|
|
#include <linux/ip.h>
|
|
|
|
#include <linux/ipv6.h>
|
|
|
|
#include <linux/icmpv6.h>
|
|
|
|
#include <linux/udp.h>
|
|
|
|
#include <net/if.h>
|
|
|
|
#include <net/if_arp.h>
|
|
|
|
#include <arpa/inet.h>
|
|
|
|
|
|
|
|
#include "passt.h"
|
|
|
|
#include "util.h"
|
passt: New design and implementation with native Layer 4 sockets
This is a reimplementation, partially building on the earlier draft,
that uses L4 sockets (SOCK_DGRAM, SOCK_STREAM) instead of SOCK_RAW,
providing L4-L2 translation functionality without requiring any
security capability.
Conceptually, this follows the design presented at:
https://gitlab.com/abologna/kubevirt-and-kvm/-/blob/master/Networking.md
The most significant novelty here comes from TCP and UDP translation
layers. In particular, the TCP state and translation logic follows
the intent of being minimalistic, without reimplementing a full TCP
stack in either direction, and synchronising as much as possible the
TCP dynamic and flows between guest and host kernel.
Another important introduction concerns addressing, port translation
and forwarding. The Layer 4 implementations now attempt to bind on
all unbound ports, in order to forward connections in a transparent
way.
While at it:
- the qemu 'tap' back-end can't be used as-is by qrap anymore,
because of explicit checks now introduced in qemu to ensure that
the corresponding file descriptor is actually a tap device. For
this reason, qrap now operates on a 'socket' back-end type,
accounting for and building the additional header reporting
frame length
- provide a demo script that sets up namespaces, addresses and
routes, and starts the daemon. A virtual machine started in the
network namespace, wrapped by qrap, will now directly interface
with passt and communicate using Layer 4 sockets provided by the
host kernel.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
2021-02-16 07:25:09 +01:00
|
|
|
#include "tap.h"
|
2020-07-21 10:48:24 +02:00
|
|
|
|
|
|
|
#define RS 133
|
|
|
|
#define RA 134
|
|
|
|
#define NS 135
|
|
|
|
#define NA 136
|
|
|
|
|
|
|
|
/**
|
|
|
|
* ndp() - Check for NDP solicitations, reply as needed
|
|
|
|
* @c: Execution context
|
|
|
|
* @len: Total L2 packet length
|
|
|
|
* @eh: Packet buffer, Ethernet header
|
|
|
|
*
|
|
|
|
* Return: 0 if not handled here, 1 if handled, -1 on failure
|
|
|
|
*/
|
|
|
|
int ndp(struct ctx *c, unsigned len, struct ethhdr *eh)
|
|
|
|
{
|
|
|
|
struct ethhdr *ehr;
|
|
|
|
struct ipv6hdr *ip6h = (struct ipv6hdr *)(eh + 1), *ip6hr;
|
|
|
|
struct icmp6hdr *ih, *ihr;
|
|
|
|
char buf[BUFSIZ] = { 0 };
|
|
|
|
uint8_t proto, *p;
|
|
|
|
|
|
|
|
ih = (struct icmp6hdr *)ipv6_l4hdr(ip6h, &proto);
|
|
|
|
if (!ih)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
if (proto != IPPROTO_ICMPV6 ||
|
|
|
|
ih->icmp6_type < RS || ih->icmp6_type > NA)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
ehr = (struct ethhdr *)buf;
|
|
|
|
ip6hr = (struct ipv6hdr *)(ehr + 1);
|
|
|
|
ihr = (struct icmp6hdr *)(ip6hr + 1);
|
|
|
|
|
|
|
|
if (ih->icmp6_type == NS) {
|
|
|
|
fprintf(stderr, "NDP: received NS, sending NA\n");
|
|
|
|
ihr->icmp6_type = NA;
|
|
|
|
ihr->icmp6_code = 0;
|
|
|
|
ihr->icmp6_router = 1;
|
|
|
|
ihr->icmp6_solicited = 1;
|
|
|
|
ihr->icmp6_override = 1;
|
|
|
|
|
|
|
|
p = (unsigned char *)(ihr + 1);
|
|
|
|
memcpy(p, &c->gw6, sizeof(c->gw6)); /* target address */
|
|
|
|
p += 16;
|
|
|
|
*p++ = 2; /* target ll */
|
|
|
|
*p++ = 1; /* length */
|
|
|
|
memcpy(p, c->mac, ETH_ALEN);
|
|
|
|
p += 6;
|
|
|
|
} else if (ih->icmp6_type == RS) {
|
|
|
|
fprintf(stderr, "NDP: received RS, sending RA\n");
|
|
|
|
ihr->icmp6_type = RA;
|
|
|
|
ihr->icmp6_code = 0;
|
|
|
|
ihr->icmp6_rt_lifetime = htons(3600);
|
|
|
|
|
|
|
|
p = (unsigned char *)(ihr + 1);
|
|
|
|
p += 8; /* reachable, retrans time */
|
|
|
|
*p++ = 3; /* prefix */
|
|
|
|
*p++ = 4; /* length */
|
|
|
|
*p++ = 64; /* prefix length */
|
|
|
|
*p++ = 0xc0; /* flags: L, A */
|
|
|
|
*(uint32_t *)p = htonl(3600); /* lifetime */
|
|
|
|
p += 4;
|
|
|
|
*(uint32_t *)p = htonl(3600); /* preferred lifetime */
|
|
|
|
p += 8;
|
|
|
|
memcpy(p, &c->addr6, 8); /* prefix */
|
|
|
|
p += 16;
|
|
|
|
|
|
|
|
*p++ = 25; /* RDNS */
|
|
|
|
*p++ = 3; /* length */
|
|
|
|
p += 2;
|
|
|
|
*(uint32_t *)p = htonl(60); /* lifetime */
|
|
|
|
p += 4;
|
|
|
|
memcpy(p, &c->dns6, 16); /* address */
|
|
|
|
p += 16;
|
|
|
|
|
|
|
|
*p++ = 1; /* source ll */
|
|
|
|
*p++ = 1; /* length */
|
|
|
|
memcpy(p, c->mac, ETH_ALEN);
|
|
|
|
p += 6;
|
|
|
|
} else {
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
len = (uintptr_t)p - (uintptr_t)ihr - sizeof(*ihr);
|
|
|
|
|
|
|
|
ip6hr->daddr = ip6h->saddr;
|
|
|
|
ip6hr->saddr = c->gw6;
|
|
|
|
ip6hr->payload_len = htons(sizeof(*ihr) + len);
|
|
|
|
ip6hr->hop_limit = IPPROTO_ICMPV6;
|
|
|
|
ihr->icmp6_cksum = 0;
|
|
|
|
ihr->icmp6_cksum = csum_ip4(ip6hr, sizeof(*ip6hr) +
|
|
|
|
sizeof(*ihr) + len);
|
|
|
|
|
|
|
|
ip6hr->version = 6;
|
|
|
|
ip6hr->nexthdr = IPPROTO_ICMPV6;
|
|
|
|
ip6hr->hop_limit = 255;
|
|
|
|
|
|
|
|
len += sizeof(*ehr) + sizeof(*ip6hr) + sizeof(*ihr);
|
|
|
|
memcpy(ehr->h_dest, eh->h_source, ETH_ALEN);
|
|
|
|
memcpy(ehr->h_source, c->mac, ETH_ALEN);
|
|
|
|
ehr->h_proto = htons(ETH_P_IPV6);
|
|
|
|
|
passt: New design and implementation with native Layer 4 sockets
This is a reimplementation, partially building on the earlier draft,
that uses L4 sockets (SOCK_DGRAM, SOCK_STREAM) instead of SOCK_RAW,
providing L4-L2 translation functionality without requiring any
security capability.
Conceptually, this follows the design presented at:
https://gitlab.com/abologna/kubevirt-and-kvm/-/blob/master/Networking.md
The most significant novelty here comes from TCP and UDP translation
layers. In particular, the TCP state and translation logic follows
the intent of being minimalistic, without reimplementing a full TCP
stack in either direction, and synchronising as much as possible the
TCP dynamic and flows between guest and host kernel.
Another important introduction concerns addressing, port translation
and forwarding. The Layer 4 implementations now attempt to bind on
all unbound ports, in order to forward connections in a transparent
way.
While at it:
- the qemu 'tap' back-end can't be used as-is by qrap anymore,
because of explicit checks now introduced in qemu to ensure that
the corresponding file descriptor is actually a tap device. For
this reason, qrap now operates on a 'socket' back-end type,
accounting for and building the additional header reporting
frame length
- provide a demo script that sets up namespaces, addresses and
routes, and starts the daemon. A virtual machine started in the
network namespace, wrapped by qrap, will now directly interface
with passt and communicate using Layer 4 sockets provided by the
host kernel.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
2021-02-16 07:25:09 +01:00
|
|
|
if (tap_send(c->fd_unix, ehr, len, 0) < 0)
|
2020-07-21 10:48:24 +02:00
|
|
|
perror("NDP: send");
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|