From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from gandalf.ozlabs.org (mail.ozlabs.org [IPv6:2404:9400:2221:ea00::3]) by passt.top (Postfix) with ESMTPS id B48425A026F for ; Mon, 24 Jul 2023 08:09:48 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gibson.dropbear.id.au; s=201602; t=1690178979; bh=3fmwIyBiM1V/G8glZX3REBS71X5ywBKCeUVbNLVU2+E=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=br/AjC/Z4eAPsuGHgntrcr7fFsa5+iDoptl4Y7PseUSRsytYbejKji7LTVkZNreDh QDpY8UPRc4Nuq1/3NIig7AqWBxJwUub6VV8OsxQUge4Auj9MphtsoxErDVlQrYKmvY J798AipOOL4Q/ctTvRw5er2kovMNnLH/Af+f4Eb4= Received: by gandalf.ozlabs.org (Postfix, from userid 1007) id 4R8V9v4WQXz4wyj; Mon, 24 Jul 2023 16:09:39 +1000 (AEST) From: David Gibson To: Stefano Brivio , passt-dev@passt.top Subject: [PATCH 05/17] netlink: Explicitly pass netlink sockets to operations Date: Mon, 24 Jul 2023 16:09:24 +1000 Message-ID: <20230724060936.952659-6-david@gibson.dropbear.id.au> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230724060936.952659-1-david@gibson.dropbear.id.au> References: <20230724060936.952659-1-david@gibson.dropbear.id.au> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-ID-Hash: UVXTKCWYOTIKKSOWAY57DLQMBBEER5XD X-Message-ID-Hash: UVXTKCWYOTIKKSOWAY57DLQMBBEER5XD X-MailFrom: dgibson@gandalf.ozlabs.org X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header CC: David Gibson X-Mailman-Version: 3.3.8 Precedence: list List-Id: Development discussion and patches for passt Archived-At: Archived-At: List-Archive: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: All the netlink operations currently implicitly use one of the two global netlink sockets, sometimes depending on an 'ns' parameter. Change them all to explicitly take the socket to use (or two sockets to use in the case of the *_dup() functions). As well as making these functions strictly more general, it makes the callers easier to follow because we're passing a socket variable with a name rather than an unexplained '0' or '1' for the ns parameter. Signed-off-by: David Gibson --- conf.c | 15 ++++----- netlink.c | 94 ++++++++++++++++++++++++++++++++----------------------- netlink.h | 28 ++++++++++------- pasta.c | 35 ++++++++++++--------- 4 files changed, 100 insertions(+), 72 deletions(-) diff --git a/conf.c b/conf.c index 66958d4..2e6e03f 100644 --- a/conf.c +++ b/conf.c @@ -640,7 +640,7 @@ static unsigned int conf_ip4(unsigned int ifi, struct ip4_ctx *ip4, unsigned char *mac) { if (!ifi) - ifi = nl_get_ext_if(AF_INET); + ifi = nl_get_ext_if(nl_sock, AF_INET); if (!ifi) { warn("No external routable interface for IPv4"); @@ -648,10 +648,11 @@ static unsigned int conf_ip4(unsigned int ifi, } if (IN4_IS_ADDR_UNSPECIFIED(&ip4->gw)) - nl_route_get_def(ifi, AF_INET, &ip4->gw); + nl_route_get_def(nl_sock, ifi, AF_INET, &ip4->gw); if (IN4_IS_ADDR_UNSPECIFIED(&ip4->addr)) - nl_addr_get(ifi, AF_INET, &ip4->addr, &ip4->prefix_len, NULL); + nl_addr_get(nl_sock, ifi, AF_INET, + &ip4->addr, &ip4->prefix_len, NULL); if (!ip4->prefix_len) { in_addr_t addr = ntohl(ip4->addr.s_addr); @@ -668,7 +669,7 @@ static unsigned int conf_ip4(unsigned int ifi, memcpy(&ip4->addr_seen, &ip4->addr, sizeof(ip4->addr_seen)); if (MAC_IS_ZERO(mac)) - nl_link_get_mac(0, ifi, mac); + nl_link_get_mac(nl_sock, ifi, mac); if (IN4_IS_ADDR_UNSPECIFIED(&ip4->addr) || MAC_IS_ZERO(mac)) @@ -691,7 +692,7 @@ static unsigned int conf_ip6(unsigned int ifi, int prefix_len = 0; if (!ifi) - ifi = nl_get_ext_if(AF_INET6); + ifi = nl_get_ext_if(nl_sock, AF_INET6); if (!ifi) { warn("No external routable interface for IPv6"); @@ -699,9 +700,9 @@ static unsigned int conf_ip6(unsigned int ifi, } if (IN6_IS_ADDR_UNSPECIFIED(&ip6->gw)) - nl_route_get_def(ifi, AF_INET6, &ip6->gw); + nl_route_get_def(nl_sock, ifi, AF_INET6, &ip6->gw); - nl_addr_get(ifi, AF_INET6, + nl_addr_get(nl_sock, ifi, AF_INET6, IN6_IS_ADDR_UNSPECIFIED(&ip6->addr) ? &ip6->addr : NULL, &prefix_len, &ip6->addr_ll); diff --git a/netlink.c b/netlink.c index 75d5988..72044cd 100644 --- a/netlink.c +++ b/netlink.c @@ -38,8 +38,8 @@ #define NLBUFSIZ (8192 * sizeof(struct nlmsghdr)) /* See netlink(7) */ /* Socket in init, in target namespace, sequence (just needs to be monotonic) */ -static int nl_sock = -1; -static int nl_sock_ns = -1; +int nl_sock = -1; +int nl_sock_ns = -1; static int nl_seq; /** @@ -98,17 +98,17 @@ fail: /** * nl_req() - Send netlink request and read response - * @ns: Use netlink socket in namespace + * @s: Netlink socket * @buf: Buffer for response (at least NLBUFSIZ long) * @req: Request with netlink header * @len: Request length * * Return: received length on success, negative error code on failure */ -static int nl_req(int ns, char *buf, const void *req, ssize_t len) +static int nl_req(int s, char *buf, const void *req, ssize_t len) { - int s = ns ? nl_sock_ns : nl_sock, done = 0; char flush[NLBUFSIZ]; + int done = 0; ssize_t n; while (!done && (n = recv(s, flush, sizeof(flush), MSG_DONTWAIT)) > 0) { @@ -133,12 +133,13 @@ static int nl_req(int ns, char *buf, const void *req, ssize_t len) /** * nl_get_ext_if() - Get interface index supporting IP version being probed + * @s: Netlink socket * @af: Address family (AF_INET or AF_INET6) to look for connectivity * for. * * Return: interface index, 0 if not found */ -unsigned int nl_get_ext_if(sa_family_t af) +unsigned int nl_get_ext_if(int s, sa_family_t af) { struct { struct nlmsghdr nlh; struct rtmsg rtm; } req = { .nlh.nlmsg_type = RTM_GETROUTE, @@ -157,7 +158,7 @@ unsigned int nl_get_ext_if(sa_family_t af) ssize_t n; size_t na; - if ((n = nl_req(0, buf, &req, sizeof(req))) < 0) + if ((n = nl_req(s, buf, &req, sizeof(req))) < 0) return 0; nh = (struct nlmsghdr *)buf; @@ -186,11 +187,12 @@ unsigned int nl_get_ext_if(sa_family_t af) /** * nl_route_get_def() - Get default route for given interface and address family + * @s: Netlink socket * @ifi: Interface index * @af: Address family * @gw: Default gateway to fill on NL_GET */ -void nl_route_get_def(unsigned int ifi, sa_family_t af, void *gw) +void nl_route_get_def(int s, unsigned int ifi, sa_family_t af, void *gw) { struct req_t { struct nlmsghdr nlh; @@ -216,7 +218,7 @@ void nl_route_get_def(unsigned int ifi, sa_family_t af, void *gw) char buf[NLBUFSIZ]; ssize_t n; - if ((n = nl_req(0, buf, &req, req.nlh.nlmsg_len)) < 0) + if ((n = nl_req(s, buf, &req, req.nlh.nlmsg_len)) < 0) return; for (nh = (struct nlmsghdr *)buf; @@ -245,11 +247,12 @@ void nl_route_get_def(unsigned int ifi, sa_family_t af, void *gw) /** * nl_route_set_def() - Set default route for given interface and address family + * @s: Netlink socket * @ifi: Interface index in target namespace * @af: Address family * @gw: Default gateway to set */ -void nl_route_set_def(unsigned int ifi, sa_family_t af, void *gw) +void nl_route_set_def(int s, unsigned int ifi, sa_family_t af, void *gw) { struct req_t { struct nlmsghdr nlh; @@ -314,16 +317,19 @@ void nl_route_set_def(unsigned int ifi, sa_family_t af, void *gw) req.set.r4.rta_gw.rta_len = rta_len; } - nl_req(1, buf, &req, req.nlh.nlmsg_len); + nl_req(s, buf, &req, req.nlh.nlmsg_len); } /** * nl_route_dup() - Copy routes for given interface and address family - * @ifi: Interface index in outer network namespace - * @ifi_ns: Interface index in target namespace for NL_SET, NL_DUP + * @s_src: Netlink socket in source namespace + * @ifi_src: Source interface index + * @s_dst: Netlink socket in destination namespace + * @ifi_dst: Interface index in destination namespace * @af: Address family */ -void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) +void nl_route_dup(int s_src, unsigned int ifi_src, + int s_dst, unsigned int ifi_dst, sa_family_t af) { struct req_t { struct nlmsghdr nlh; @@ -343,7 +349,7 @@ void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) .rta.rta_type = RTA_OIF, .rta.rta_len = RTA_LENGTH(sizeof(unsigned int)), - .ifi = ifi, + .ifi = ifi_src, }; char buf[NLBUFSIZ], resp[NLBUFSIZ]; unsigned dup_routes = 0; @@ -351,7 +357,7 @@ void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) struct nlmsghdr *nh; unsigned i; - if ((n = nl_req(0, buf, &req, req.nlh.nlmsg_len)) < 0) + if ((n = nl_req(s_src, buf, &req, req.nlh.nlmsg_len)) < 0) return; nlmsgs_size = n; @@ -376,7 +382,7 @@ void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) for (rta = RTM_RTA(rtm), na = RTM_PAYLOAD(nh); RTA_OK(rta, na); rta = RTA_NEXT(rta, na)) { if (rta->rta_type == RTA_OIF) - *(unsigned int *)RTA_DATA(rta) = ifi_ns; + *(unsigned int *)RTA_DATA(rta) = ifi_dst; } } @@ -390,19 +396,20 @@ void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) * need to calculate dependencies: let the kernel do that. */ for (i = 0; i < dup_routes; i++) - nl_req(1, resp, nh, nlmsgs_size); + nl_req(s_dst, resp, nh, nlmsgs_size); } /** * nl_addr_get() - Get IP address for given interface and address family + * @s: Netlink socket * @ifi: Interface index in outer network namespace * @af: Address family * @addr: Global address to fill * @prefix_len: Mask or prefix length, to fill (for IPv4) * @addr_l: Link-scoped address to fill (for IPv6) */ -void nl_addr_get(unsigned int ifi, sa_family_t af, void *addr, - int *prefix_len, void *addr_l) +void nl_addr_get(int s, unsigned int ifi, sa_family_t af, + void *addr, int *prefix_len, void *addr_l) { struct req_t { struct nlmsghdr nlh; @@ -420,7 +427,7 @@ void nl_addr_get(unsigned int ifi, sa_family_t af, void *addr, char buf[NLBUFSIZ]; ssize_t n; - if ((n = nl_req(0, buf, &req, req.nlh.nlmsg_len)) < 0) + if ((n = nl_req(s, buf, &req, req.nlh.nlmsg_len)) < 0) return; for (nh = (struct nlmsghdr *)buf; @@ -458,12 +465,14 @@ void nl_addr_get(unsigned int ifi, sa_family_t af, void *addr, /** * nl_add_set() - Set IP addresses for given interface and address family + * @s: Netlink socket * @ifi: Interface index * @af: Address family * @addr: Global address to set * @prefix_len: Mask or prefix length to set */ -void nl_addr_set(unsigned int ifi, sa_family_t af, void *addr, int prefix_len) +void nl_addr_set(int s, unsigned int ifi, sa_family_t af, + void *addr, int prefix_len) { struct req_t { struct nlmsghdr nlh; @@ -524,16 +533,19 @@ void nl_addr_set(unsigned int ifi, sa_family_t af, void *addr, int prefix_len) req.set.a4.rta_a.rta_type = IFA_ADDRESS; } - nl_req(1, buf, &req, req.nlh.nlmsg_len); + nl_req(s, buf, &req, req.nlh.nlmsg_len); } /** * nl_addr_dup() - Copy IP addresses for given interface and address family - * @ifi: Interface index in outer network namespace - * @ifi_ns: Interface index in target namespace + * @s_src: Netlink socket in source network namespace + * @ifi_src: Interface index in source network namespace + * @s_dst: Netlink socket in destination network namespace + * @ifi_dst: Interface index in destination namespace * @af: Address family */ -void nl_addr_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) +void nl_addr_dup(int s_src, unsigned int ifi_src, + int s_dst, unsigned int ifi_dst, sa_family_t af) { struct req_t { struct nlmsghdr nlh; @@ -545,14 +557,14 @@ void nl_addr_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) .nlh.nlmsg_seq = nl_seq++, .ifa.ifa_family = af, - .ifa.ifa_index = ifi, + .ifa.ifa_index = ifi_src, .ifa.ifa_prefixlen = 0, }; char buf[NLBUFSIZ], resp[NLBUFSIZ]; ssize_t n, nlmsgs_size; struct nlmsghdr *nh; - if ((n = nl_req(0, buf, &req, sizeof(req))) < 0) + if ((n = nl_req(s_src, buf, &req, sizeof(req))) < 0) return; nlmsgs_size = n; @@ -574,12 +586,13 @@ void nl_addr_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) ifa = (struct ifaddrmsg *)NLMSG_DATA(nh); - if (ifa->ifa_scope == RT_SCOPE_LINK || ifa->ifa_index != ifi) { + if (ifa->ifa_scope == RT_SCOPE_LINK || + ifa->ifa_index != ifi_src) { ifa->ifa_family = AF_UNSPEC; continue; } - ifa->ifa_index = ifi_ns; + ifa->ifa_index = ifi_dst; for (rta = IFA_RTA(ifa), na = RTM_PAYLOAD(nh); RTA_OK(rta, na); rta = RTA_NEXT(rta, na)) { @@ -588,16 +601,16 @@ void nl_addr_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) } } - nl_req(1, resp, buf, nlmsgs_size); + nl_req(s_dst, resp, buf, nlmsgs_size); } /** * nl_link_get_mac() - Get link MAC address - * @ns: Use netlink socket in namespace + * @s: Netlink socket * @ifi: Interface index * @mac: Fill with current MAC address */ -void nl_link_get_mac(int ns, unsigned int ifi, void *mac) +void nl_link_get_mac(int s, unsigned int ifi, void *mac) { struct req_t { struct nlmsghdr nlh; @@ -614,10 +627,10 @@ void nl_link_get_mac(int ns, unsigned int ifi, void *mac) char buf[NLBUFSIZ]; ssize_t n; - n = nl_req(ns, buf, &req, sizeof(req)); + n = nl_req(s, buf, &req, sizeof(req)); if (n < 0) return; - + for (nh = (struct nlmsghdr *)buf; NLMSG_OK(nh, n) && nh->nlmsg_type != NLMSG_DONE; nh = NLMSG_NEXT(nh, n)) { @@ -642,11 +655,12 @@ void nl_link_get_mac(int ns, unsigned int ifi, void *mac) /** * nl_link_set_mac() - Set link MAC address + * @s: Netlink socket * @ns: Use netlink socket in namespace * @ifi: Interface index * @mac: MAC address to set */ -void nl_link_set_mac(int ns, unsigned int ifi, void *mac) +void nl_link_set_mac(int s, unsigned int ifi, void *mac) { struct req_t { struct nlmsghdr nlh; @@ -667,16 +681,16 @@ void nl_link_set_mac(int ns, unsigned int ifi, void *mac) memcpy(req.mac, mac, ETH_ALEN); - nl_req(ns, buf, &req, sizeof(req)); + nl_req(s, buf, &req, sizeof(req)); } /** * nl_link_up() - Bring link up - * @ns: Use netlink socket in namespace + * @s: Netlink socket * @ifi: Interface index * @mtu: If non-zero, set interface MTU */ -void nl_link_up(int ns, unsigned int ifi, int mtu) +void nl_link_up(int s, unsigned int ifi, int mtu) { struct req_t { struct nlmsghdr nlh; @@ -702,5 +716,5 @@ void nl_link_up(int ns, unsigned int ifi, int mtu) /* Shorten request to drop MTU attribute */ req.nlh.nlmsg_len = offsetof(struct req_t, rta); - nl_req(ns, buf, &req, req.nlh.nlmsg_len); + nl_req(s, buf, &req, req.nlh.nlmsg_len); } diff --git a/netlink.h b/netlink.h index 36bbf9f..5ca17c6 100644 --- a/netlink.h +++ b/netlink.h @@ -6,17 +6,23 @@ #ifndef NETLINK_H #define NETLINK_H +extern int nl_sock; +extern int nl_sock_ns; + void nl_sock_init(const struct ctx *c, bool ns); -unsigned int nl_get_ext_if(sa_family_t af); -void nl_route_get_def(unsigned int ifi, sa_family_t af, void *gw); -void nl_route_set_def(unsigned int ifi, sa_family_t af, void *gw); -void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af); -void nl_addr_get(unsigned int ifi, sa_family_t af, void *addr, - int *prefix_len, void *addr_l); -void nl_addr_set(unsigned int ifi, sa_family_t af, void *addr, int prefix_len); -void nl_addr_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af); -void nl_link_get_mac(int ns, unsigned int ifi, void *mac); -void nl_link_set_mac(int ns, unsigned int ifi, void *mac); -void nl_link_up(int ns, unsigned int ifi, int mtu); +unsigned int nl_get_ext_if(int s, sa_family_t af); +void nl_route_get_def(int s, unsigned int ifi, sa_family_t af, void *gw); +void nl_route_set_def(int s, unsigned int ifi, sa_family_t af, void *gw); +void nl_route_dup(int s_src, unsigned int ifi_src, + int s_dst, unsigned int ifi_dst, sa_family_t af); +void nl_addr_get(int s, unsigned int ifi, sa_family_t af, + void *addr, int *prefix_len, void *addr_l); +void nl_addr_set(int s, unsigned int ifi, sa_family_t af, + void *addr, int prefix_len); +void nl_addr_dup(int s_src, unsigned int ifi_src, + int s_dst, unsigned int ifi_dst, sa_family_t af); +void nl_link_get_mac(int s, unsigned int ifi, void *mac); +void nl_link_set_mac(int s, unsigned int ifi, void *mac); +void nl_link_up(int s, unsigned int ifi, int mtu); #endif /* NETLINK_H */ diff --git a/pasta.c b/pasta.c index 14ecc71..3380475 100644 --- a/pasta.c +++ b/pasta.c @@ -272,42 +272,49 @@ void pasta_start_ns(struct ctx *c, uid_t uid, gid_t gid, */ void pasta_ns_conf(struct ctx *c) { - nl_link_up(1, 1 /* lo */, 0); + nl_link_up(nl_sock_ns, 1 /* lo */, 0); /* Get or set guest MAC */ if (MAC_IS_ZERO(c->mac_guest)) - nl_link_get_mac(1, c->pasta_ifi, c->mac_guest); + nl_link_get_mac(nl_sock_ns, c->pasta_ifi, c->mac_guest); else - nl_link_set_mac(1, c->pasta_ifi, c->mac_guest); + nl_link_set_mac(nl_sock_ns, c->pasta_ifi, c->mac_guest); if (c->pasta_conf_ns) { - nl_link_up(1, c->pasta_ifi, c->mtu); + nl_link_up(nl_sock_ns, c->pasta_ifi, c->mtu); if (c->ifi4) { if (c->no_copy_addrs) - nl_addr_set(c->pasta_ifi, AF_INET, + nl_addr_set(nl_sock_ns, c->pasta_ifi, AF_INET, &c->ip4.addr, c->ip4.prefix_len); else - nl_addr_dup(c->ifi4, c->pasta_ifi, AF_INET); + nl_addr_dup(nl_sock, c->ifi4, + nl_sock_ns, c->pasta_ifi, AF_INET); if (c->no_copy_routes) - nl_route_set_def(c->pasta_ifi, AF_INET, - &c->ip4.gw); + nl_route_set_def(nl_sock_ns, c->pasta_ifi, + AF_INET, &c->ip4.gw); else - nl_route_dup(c->ifi4, c->pasta_ifi, AF_INET); + nl_route_dup(nl_sock, c->ifi4, nl_sock_ns, + c->pasta_ifi, AF_INET); } if (c->ifi6) { if (c->no_copy_addrs) - nl_addr_set(c->pasta_ifi, AF_INET6, &c->ip6.addr, 64); + nl_addr_set(nl_sock_ns, c->pasta_ifi, + AF_INET6, &c->ip6.addr, 64); else - nl_addr_dup(c->ifi4, c->pasta_ifi, AF_INET6); + nl_addr_dup(nl_sock, c->ifi4, + nl_sock_ns, c->pasta_ifi, + AF_INET6); if (c->no_copy_routes) - nl_route_set_def(c->pasta_ifi, AF_INET6, - &c->ip6.gw); + nl_route_set_def(nl_sock_ns, c->pasta_ifi, + AF_INET6, &c->ip6.gw); else - nl_route_dup(c->ifi6, c->pasta_ifi, AF_INET6); + nl_route_dup(nl_sock, c->ifi6, + nl_sock_ns, c->pasta_ifi, + AF_INET6); } } -- 2.41.0