// SPDX-License-Identifier: GPL-2.0-or-later

/* PASST - Plug A Simple Socket Transport
 *  for qemu/UNIX domain socket mode
 *
 * PASTA - Pack A Subtle Tap Abstraction
 *  for network namespace/tap device mode
 *
 * icmp.c - ICMP/ICMPv6 echo proxy
 *
 * Copyright (c) 2021 Red Hat GmbH
 * Author: Stefano Brivio <sbrivio@redhat.com>
 */

#include <errno.h>
#include <net/ethernet.h>
#include <net/if.h>
#include <netinet/in.h>
#include <netinet/ip.h>
#include <netinet/ip_icmp.h>
#include <stdio.h>
#include <limits.h>
#include <stdint.h>
#include <stddef.h>
#include <string.h>
#include <sys/epoll.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <unistd.h>
#include <time.h>

#include <linux/icmpv6.h>

#include "packet.h"
#include "util.h"
#include "passt.h"
#include "tap.h"
#include "log.h"
#include "icmp.h"
#include "flow_table.h"

#define ICMP_ECHO_TIMEOUT	60 /* s, timeout for ICMP socket activity */
#define ICMP_NUM_IDS		(1U << 16)

/* Sides of a flow as we use them for ping streams */
#define	SOCKSIDE	0
#define	TAPSIDE		1

#define PINGF(idx)		(&(FLOW(idx)->ping))

#define TAPFSIDE(pingf)		(&(pingf)->f.side[TAPSIDE])
#define SOCKFSIDE(pingf)	(&(pingf)->f.side[SOCKSIDE])

/* Indexed by ICMP echo identifier */
static struct icmp_ping_flow *icmp_id_map[IP_VERSIONS][ICMP_NUM_IDS];

/**
 * icmp_sock_handler() - Handle new data from ICMP or ICMPv6 socket
 * @c:		Execution context
 * @ref:	epoll reference
 */
void icmp_sock_handler(const struct ctx *c, union epoll_ref ref)
{
	struct icmp_ping_flow *pingf = PINGF(ref.flowside.flow);
	int af = pingf->f.type == FLOW_PING4 ? AF_INET : AF_INET6;
	const char *const pname = af == AF_INET ? "ICMP" : "ICMPv6";
	char buf[USHRT_MAX];
	union {
		struct sockaddr sa;
		struct sockaddr_in sa4;
		struct sockaddr_in6 sa6;
	} sr;
	uint16_t id = TAPFSIDE(pingf)->eport, seq;
	socklen_t sl = sizeof(sr);
	ssize_t n;

	if (c->no_icmp)
		return;

	ASSERT(pingf);

	n = recvfrom(ref.fd, buf, sizeof(buf), 0, &sr.sa, &sl);
	if (n < 0) {
		warn("%s: recvfrom() error on ping socket: %s",
		     pname, strerror(errno));
		return;
	}
	if (sr.sa.sa_family != af)
		goto unexpected;

	if (af == AF_INET) {
		struct icmphdr *ih4 = (struct icmphdr *)buf;

		if ((size_t)n < sizeof(*ih4) || ih4->type != ICMP_ECHOREPLY)
			goto unexpected;

		/* Adjust packet back to guest-side ID */
		ih4->un.echo.id = htons(id);
		seq = ntohs(ih4->un.echo.sequence);
	} else if (af == AF_INET6) {
		struct icmp6hdr *ih6 = (struct icmp6hdr *)buf;

		if ((size_t)n < sizeof(*ih6) ||
		    ih6->icmp6_type != ICMPV6_ECHO_REPLY)
			goto unexpected;

		/* Adjust packet back to guest-side ID */
		ih6->icmp6_identifier = htons(id);
		seq = ntohs(ih6->icmp6_sequence);
	} else {
		ASSERT(0);
	}

	if (c->mode == MODE_PASTA) {
		if (pingf->seq == seq)
			return;

		pingf->seq = seq;
	}

	debug("%s: echo reply to tap, ID: %"PRIu16", seq: %"PRIu16, pname,
	      id, seq);
	if (af == AF_INET) {
		const struct in_addr *saddr = inany_v4(&TAPFSIDE(pingf)->faddr);
		const struct in_addr *daddr = inany_v4(&TAPFSIDE(pingf)->eaddr);

		ASSERT(saddr && daddr); /* Must have IPv4 addresses */
		tap_icmp4_send(c, *saddr, *daddr, buf, n);
	} else if (af == AF_INET6) {
		const struct in6_addr *saddr = &TAPFSIDE(pingf)->faddr.a6;
		const struct in6_addr *daddr = &TAPFSIDE(pingf)->eaddr.a6;
		tap_icmp6_send(c, saddr, daddr, buf, n);
	}
	return;

unexpected:
	warn("%s: Unexpected packet on ping socket", pname);
}

/**
 * icmp_ping_close() - Close out and cleanup a ping flow
 * @c:		Execution context
 * @pingf:	ping flow entry to close
 */
static void icmp_ping_close(const struct ctx *c, struct icmp_ping_flow *pingf)
{
	epoll_ctl(c->epollfd, EPOLL_CTL_DEL, pingf->sock, NULL);
	close(pingf->sock);
	flow_hash_remove(c, FLOW_SIDX(pingf, TAPSIDE));

	if (pingf->f.type == FLOW_PING4)
		icmp_id_map[V4][pingf->id] = NULL;
	else
		icmp_id_map[V6][pingf->id] = NULL;
}

/**
 * icmp_ping_new() - Prepare a new ping socket for a new id
 * @c:		Execution context
 * @id_map:	id map entry of the sequence to open
 * @af:		Address family, AF_INET or AF_INET6
 * @id:		ICMP id for the new sequence
 * @saddr:	Source address
 * @daddr:	Destination address
 *
 * Return: Newly opened ping flow, or NULL on failure
 */
static struct icmp_ping_flow *icmp_ping_new(const struct ctx *c,
					    struct icmp_ping_flow **id_map,
					    int af, uint16_t id,
					    const void *saddr, const void *daddr)
{
	const char *const pname = af == AF_INET ? "ICMP" : "ICMPv6";
	uint8_t flowtype = af == AF_INET ? FLOW_PING4 : FLOW_PING6;
	union epoll_ref ref = { .type = EPOLL_TYPE_PING };
	union flow *flow = flow_alloc();
	struct icmp_ping_flow *pingf;
	const void *bind_addr;
	const char *bind_if;
	int s;

	if (!flow)
		return NULL;

	if (af == AF_INET) {
		bind_addr = &c->ip4.addr_out;
		bind_if = c->ip4.ifname_out;
	} else {
		bind_addr = &c->ip6.addr_out;
		bind_if = c->ip6.ifname_out;
	}

	ref.flowside = FLOW_SIDX(flow, SOCKSIDE);
	s = sock_l4(c, af, flow_proto[flowtype], bind_addr, bind_if, 0, ref.data);

	if (s < 0) {
		warn("Cannot open \"ping\" socket. You might need to:");
		warn("  sysctl -w net.ipv4.ping_group_range=\"0 2147483647\"");
		warn("...echo requests/replies will fail.");
		goto cancel;
	}

	if (s > FD_REF_MAX)
		goto cancel;

	pingf = &flow->ping;
	pingf->f.type = flowtype;
	pingf->seq = -1;
	pingf->sock = s;
	pingf->id = id;

	*id_map = pingf;

	debug("%s: new socket %i for echo ID %"PRIu16, pname, s, id);

	flowside_from_af(TAPFSIDE(pingf), PIF_TAP, af, daddr, id, saddr, id);
	FLOW_NEW_DBG(pingf, TAPSIDE);
	flow_hash_insert(c, FLOW_SIDX(pingf, TAPSIDE));

	return pingf;

cancel:
	if (s >= 0)
		close(s);
	flow_alloc_cancel(flow);
	return NULL;
}

/**
 * icmp_tap_handler() - Handle packets from tap
 * @c:		Execution context
 * @pif:	pif on which the packet is arriving
 * @af:		Address family, AF_INET or AF_INET6
 * @saddr:	Source address
 * @daddr:	Destination address
 * @p:		Packet pool, single packet with ICMP/ICMPv6 header
 * @now:	Current timestamp
 *
 * Return: count of consumed packets (always 1, even if malformed)
 */
int icmp_tap_handler(const struct ctx *c, uint8_t pif, int af,
		     const void *saddr, const void *daddr,
		     const struct pool *p, const struct timespec *now)
{
	const char *const pname = af == AF_INET ? "ICMP" : "ICMPv6";
	union {
		struct sockaddr sa;
		struct sockaddr_in sa4;
		struct sockaddr_in6 sa6;
	} sa = { .sa.sa_family = af };
	const socklen_t sl = af == AF_INET ? sizeof(sa.sa4) : sizeof(sa.sa6);
	struct icmp_ping_flow *pingf, **id_map;
	union flow *flow;
	uint16_t id, seq;
	uint8_t proto;
	size_t plen;
	void *pkt;

	(void)saddr;
	(void)pif;

	pkt = packet_get(p, 0, 0, 0, &plen);
	if (!pkt)
		return 1;

	if (af == AF_INET) {
		struct icmphdr *ih = (struct icmphdr *)pkt;

		if (plen < sizeof(*ih))
			return 1;

		if (ih->type != ICMP_ECHO)
			return 1;

		proto = IPPROTO_ICMP;
		id = ntohs(ih->un.echo.id);
		id_map = &icmp_id_map[V4][id];
		seq = ntohs(ih->un.echo.sequence);
		sa.sa4.sin_addr = *(struct in_addr *)daddr;
	} else if (af == AF_INET6) {
		struct icmp6hdr *ih = (struct icmp6hdr *)pkt;

		if (plen < sizeof(*ih))
			return 1;

		if (ih->icmp6_type != ICMPV6_ECHO_REQUEST)
			return 1;

		proto = IPPROTO_ICMPV6;
		id = ntohs(ih->icmp6_identifier);
		id_map = &icmp_id_map[V6][id];
		seq = ntohs(ih->icmp6_sequence);
		sa.sa6.sin6_addr = *(struct in6_addr *)daddr;
		sa.sa6.sin6_scope_id = c->ifi6;
	} else {
		ASSERT(0);
	}

	flow = flow_at_sidx(flow_hash_lookup(c, proto, PIF_TAP,
					     af, saddr, daddr, id, id));

	if (flow)
		pingf = &flow->ping;
	else if (!(pingf = icmp_ping_new(c, id_map, af, id, saddr, daddr)))
		return 1;

	ASSERT(flow_proto[pingf->f.type] == proto);
	pingf->ts = now->tv_sec;

	if (sendto(pingf->sock, pkt, plen, MSG_NOSIGNAL, &sa.sa, sl) < 0) {
		debug("%s: failed to relay request to socket: %s",
		      pname, strerror(errno));
		if (flow)
			goto cancel;
	}

	debug("%s: echo request to socket, ID: %"PRIu16", seq: %"PRIu16,
	      pname, id, seq);

	if (!flow)
		/* Nothing more to do for an existing flow */
		return 1;

	/* We need to wait until after the sendto() to fill in the SOCKSIDE
	 * information, so that we can find out the host side id the kernel
	 * assigned.  If there's no bind address specified, this will still have
	 * 0.0.0.0 or :: as the host side forwarding address.  There's not
	 * really anything we can do to fill that in, which means we can never
	 * insert the SOCKSIDE of a ping flow into the hash table.
	 */
	if (flowside_from_sock(SOCKFSIDE(pingf), PIF_HOST, pingf->sock,
			       NULL, &sa) < 0) {
		err("%s: Failed to get local name for outgoing ping socket",
		    pname);
		goto cancel;
	}
	/* We want the id as the "port" on both sides */
	SOCKFSIDE(pingf)->eport = SOCKFSIDE(pingf)->fport;

	FLOW_FWD_DBG(pingf, SOCKSIDE);

	return 1;

cancel:
	/* Something went wrong, back out creation of the flow */
	icmp_ping_close(c, pingf);
	flow_alloc_cancel(flow);
	return 1;
}

/**
 * icmp_ping_timer() - Handler for timed events related to a given flow
 * @c:		Execution context
 * @flow:	flow table entry to check for timeout
 * @now:	Current timestamp
 *
 * Return: true if the flow is ready to free, false otherwise
 */
bool icmp_ping_timer(const struct ctx *c, union flow *flow,
		     const struct timespec *now)
{
	struct icmp_ping_flow *pingf = &flow->ping;

	if (now->tv_sec - pingf->ts <= ICMP_ECHO_TIMEOUT)
		return false;

	icmp_ping_close(c, pingf);
	return true;
}