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 8F6EC5A026F for ; Mon, 24 Jul 2023 08:09:43 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gibson.dropbear.id.au; s=201602; t=1690178979; bh=OLph3Pf6Y0fx5YP3H8ZRDHiXJ1ZtMoFjLMQ0elVAAX0=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=PCOWMJo0JNAhTzUfscd7K4XoagVxUEXHpZafJDcuQznneIK8rubFZEtNXl6CnFt8e 0w/ffp/DxT+DFFiM1IigMIw/ss/oNkiHqJI4uUhK8fnEVAajKBD/tzILg1HbuWZQFT oyohVui6T8vkik4cDQywyb1Ggpx2Rj8o8x50nF+o= Received: by gandalf.ozlabs.org (Postfix, from userid 1007) id 4R8V9v4FD4z4wyg; Mon, 24 Jul 2023 16:09:39 +1000 (AEST) From: David Gibson To: Stefano Brivio , passt-dev@passt.top Subject: [PATCH 03/17] netlink: Split nl_route() into separate operation functions Date: Mon, 24 Jul 2023 16:09:22 +1000 Message-ID: <20230724060936.952659-4-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: C3AVP2T5RJKLG7EMQDBLUGD77RICMSEN X-Message-ID-Hash: C3AVP2T5RJKLG7EMQDBLUGD77RICMSEN 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: nl_route() can perform 3 quite different operations based on the 'op' parameter. Split this into separate functions for each one. This requires more lines of code, but makes the internal logic of each operation much easier to follow. Signed-off-by: David Gibson --- conf.c | 4 +- netlink.c | 238 ++++++++++++++++++++++++++++++++++-------------------- netlink.h | 11 +-- pasta.c | 16 ++-- 4 files changed, 164 insertions(+), 105 deletions(-) diff --git a/conf.c b/conf.c index 2057028..66958d4 100644 --- a/conf.c +++ b/conf.c @@ -648,7 +648,7 @@ static unsigned int conf_ip4(unsigned int ifi, } if (IN4_IS_ADDR_UNSPECIFIED(&ip4->gw)) - nl_route(NL_GET, ifi, 0, AF_INET, &ip4->gw); + nl_route_get_def(ifi, AF_INET, &ip4->gw); if (IN4_IS_ADDR_UNSPECIFIED(&ip4->addr)) nl_addr_get(ifi, AF_INET, &ip4->addr, &ip4->prefix_len, NULL); @@ -699,7 +699,7 @@ static unsigned int conf_ip6(unsigned int ifi, } if (IN6_IS_ADDR_UNSPECIFIED(&ip6->gw)) - nl_route(NL_GET, ifi, 0, AF_INET6, &ip6->gw); + nl_route_get_def(ifi, AF_INET6, &ip6->gw); nl_addr_get(ifi, AF_INET6, IN6_IS_ADDR_UNSPECIFIED(&ip6->addr) ? &ip6->addr : NULL, diff --git a/netlink.c b/netlink.c index 269d738..346eb3a 100644 --- a/netlink.c +++ b/netlink.c @@ -185,15 +185,71 @@ unsigned int nl_get_ext_if(sa_family_t af) } /** - * nl_route() - Get/set/copy routes for given interface and address family - * @op: Requested operation - * @ifi: Interface index in outer network namespace - * @ifi_ns: Interface index in target namespace for NL_SET, NL_DUP + * nl_route_get_def() - Get default route for given interface and address family + * @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) +{ + struct req_t { + struct nlmsghdr nlh; + struct rtmsg rtm; + struct rtattr rta; + unsigned int ifi; + } req = { + .nlh.nlmsg_type = RTM_GETROUTE, + .nlh.nlmsg_len = sizeof(req), + .nlh.nlmsg_flags = NLM_F_REQUEST | NLM_F_DUMP, + .nlh.nlmsg_seq = nl_seq++, + + .rtm.rtm_family = af, + .rtm.rtm_table = RT_TABLE_MAIN, + .rtm.rtm_scope = RT_SCOPE_UNIVERSE, + .rtm.rtm_type = RTN_UNICAST, + + .rta.rta_type = RTA_OIF, + .rta.rta_len = RTA_LENGTH(sizeof(unsigned int)), + .ifi = ifi, + }; + struct nlmsghdr *nh; + char buf[NLBUFSIZ]; + ssize_t n; + + if ((n = nl_req(0, buf, &req, req.nlh.nlmsg_len)) < 0) + return; + + for (nh = (struct nlmsghdr *)buf; + NLMSG_OK(nh, n) && nh->nlmsg_type != NLMSG_DONE; + nh = NLMSG_NEXT(nh, n)) { + struct rtmsg *rtm = (struct rtmsg *)NLMSG_DATA(nh); + struct rtattr *rta; + size_t na; + + if (nh->nlmsg_type != RTM_NEWROUTE) + continue; + + if (rtm->rtm_dst_len) + continue; + + for (rta = RTM_RTA(rtm), na = RTM_PAYLOAD(nh); RTA_OK(rta, na); + rta = RTA_NEXT(rta, na)) { + if (rta->rta_type != RTA_GATEWAY) + continue; + + memcpy(gw, RTA_DATA(rta), RTA_PAYLOAD(rta)); + return; + } + } +} + +/** + * nl_route_set_def() - Set default route for given interface and address family + * @ifi: Interface index in target namespace * @af: Address family - * @gw: Default gateway to fill on NL_GET, to set on NL_SET + * @gw: Default gateway to set */ -void nl_route(enum nl_op op, unsigned int ifi, unsigned int ifi_ns, - sa_family_t af, void *gw) +void nl_route_set_def(unsigned int ifi, sa_family_t af, void *gw) { struct req_t { struct nlmsghdr nlh; @@ -215,122 +271,126 @@ void nl_route(enum nl_op op, unsigned int ifi, unsigned int ifi_ns, } r4; } set; } req = { - .nlh.nlmsg_type = op == NL_SET ? RTM_NEWROUTE : RTM_GETROUTE, - .nlh.nlmsg_flags = NLM_F_REQUEST, + .nlh.nlmsg_type = RTM_NEWROUTE, + .nlh.nlmsg_flags = NLM_F_REQUEST | NLM_F_ACK | + NLM_F_CREATE | NLM_F_EXCL, .nlh.nlmsg_seq = nl_seq++, .rtm.rtm_family = af, .rtm.rtm_table = RT_TABLE_MAIN, .rtm.rtm_scope = RT_SCOPE_UNIVERSE, .rtm.rtm_type = RTN_UNICAST, + .rtm.rtm_protocol = RTPROT_BOOT, .rta.rta_type = RTA_OIF, .rta.rta_len = RTA_LENGTH(sizeof(unsigned int)), - .ifi = op == NL_SET ? ifi_ns : ifi, + .ifi = ifi, }; - unsigned dup_routes = 0; - ssize_t n, nlmsgs_size; - struct nlmsghdr *nh; - struct rtattr *rta; char buf[NLBUFSIZ]; - struct rtmsg *rtm; - size_t na; - - if (op == NL_SET) { - if (af == AF_INET6) { - size_t rta_len = RTA_LENGTH(sizeof(req.set.r6.d)); - req.nlh.nlmsg_len = offsetof(struct req_t, set.r6) - + sizeof(req.set.r6); + if (af == AF_INET6) { + size_t rta_len = RTA_LENGTH(sizeof(req.set.r6.d)); - req.set.r6.rta_dst.rta_type = RTA_DST; - req.set.r6.rta_dst.rta_len = rta_len; + req.nlh.nlmsg_len = offsetof(struct req_t, set.r6) + + sizeof(req.set.r6); - memcpy(&req.set.r6.a, gw, sizeof(req.set.r6.a)); - req.set.r6.rta_gw.rta_type = RTA_GATEWAY; - req.set.r6.rta_gw.rta_len = rta_len; - } else { - size_t rta_len = RTA_LENGTH(sizeof(req.set.r4.d)); + req.set.r6.rta_dst.rta_type = RTA_DST; + req.set.r6.rta_dst.rta_len = rta_len; - req.nlh.nlmsg_len = offsetof(struct req_t, set.r4) - + sizeof(req.set.r4); + memcpy(&req.set.r6.a, gw, sizeof(req.set.r6.a)); + req.set.r6.rta_gw.rta_type = RTA_GATEWAY; + req.set.r6.rta_gw.rta_len = rta_len; + } else { + size_t rta_len = RTA_LENGTH(sizeof(req.set.r4.d)); - req.set.r4.rta_dst.rta_type = RTA_DST; - req.set.r4.rta_dst.rta_len = rta_len; + req.nlh.nlmsg_len = offsetof(struct req_t, set.r4) + + sizeof(req.set.r4); - req.set.r4.a = *(uint32_t *)gw; - req.set.r4.rta_gw.rta_type = RTA_GATEWAY; - req.set.r4.rta_gw.rta_len = rta_len; - } + req.set.r4.rta_dst.rta_type = RTA_DST; + req.set.r4.rta_dst.rta_len = rta_len; - req.rtm.rtm_protocol = RTPROT_BOOT; - req.nlh.nlmsg_flags |= NLM_F_ACK | NLM_F_EXCL | NLM_F_CREATE; - } else { - req.nlh.nlmsg_len = offsetof(struct req_t, set.r6); - req.nlh.nlmsg_flags |= NLM_F_DUMP; + req.set.r4.a = *(uint32_t *)gw; + req.set.r4.rta_gw.rta_type = RTA_GATEWAY; + req.set.r4.rta_gw.rta_len = rta_len; } - if ((n = nl_req(op == NL_SET, buf, &req, req.nlh.nlmsg_len)) < 0) - return; + nl_req(1, buf, &req, req.nlh.nlmsg_len); +} - if (op == NL_SET) +/** + * 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 + * @af: Address family + */ +void nl_route_dup(unsigned int ifi, unsigned int ifi_ns, sa_family_t af) +{ + struct req_t { + struct nlmsghdr nlh; + struct rtmsg rtm; + struct rtattr rta; + unsigned int ifi; + } req = { + .nlh.nlmsg_type = RTM_GETROUTE, + .nlh.nlmsg_len = sizeof(req), + .nlh.nlmsg_flags = NLM_F_REQUEST | NLM_F_DUMP, + .nlh.nlmsg_seq = nl_seq++, + + .rtm.rtm_family = af, + .rtm.rtm_table = RT_TABLE_MAIN, + .rtm.rtm_scope = RT_SCOPE_UNIVERSE, + .rtm.rtm_type = RTN_UNICAST, + + .rta.rta_type = RTA_OIF, + .rta.rta_len = RTA_LENGTH(sizeof(unsigned int)), + .ifi = ifi, + }; + char buf[NLBUFSIZ], resp[NLBUFSIZ]; + unsigned dup_routes = 0; + ssize_t n, nlmsgs_size; + struct nlmsghdr *nh; + unsigned i; + + if ((n = nl_req(0, buf, &req, req.nlh.nlmsg_len)) < 0) return; - nh = (struct nlmsghdr *)buf; nlmsgs_size = n; - for ( ; NLMSG_OK(nh, n); nh = NLMSG_NEXT(nh, n)) { - if (nh->nlmsg_type != RTM_NEWROUTE) - goto next; - - if (op == NL_DUP) { - nh->nlmsg_seq = nl_seq++; - nh->nlmsg_pid = 0; - nh->nlmsg_flags &= ~NLM_F_DUMP_FILTERED; - nh->nlmsg_flags |= NLM_F_REQUEST | NLM_F_ACK | - NLM_F_CREATE; - dup_routes++; - } + for (nh = (struct nlmsghdr *)buf; + NLMSG_OK(nh, n) && nh->nlmsg_type != NLMSG_DONE; + nh = NLMSG_NEXT(nh, n)) { + struct rtmsg *rtm = (struct rtmsg *)NLMSG_DATA(nh); + struct rtattr *rta; + size_t na; - rtm = (struct rtmsg *)NLMSG_DATA(nh); - if (op == NL_GET && rtm->rtm_dst_len) + if (nh->nlmsg_type != RTM_NEWROUTE) continue; + nh->nlmsg_seq = nl_seq++; + nh->nlmsg_pid = 0; + nh->nlmsg_flags &= ~NLM_F_DUMP_FILTERED; + nh->nlmsg_flags |= NLM_F_REQUEST | NLM_F_ACK | + NLM_F_CREATE; + dup_routes++; + for (rta = RTM_RTA(rtm), na = RTM_PAYLOAD(nh); RTA_OK(rta, na); rta = RTA_NEXT(rta, na)) { - if (op == NL_GET) { - if (rta->rta_type != RTA_GATEWAY) - continue; - - memcpy(gw, RTA_DATA(rta), RTA_PAYLOAD(rta)); - return; - } - - if (op == NL_DUP && rta->rta_type == RTA_OIF) + if (rta->rta_type == RTA_OIF) *(unsigned int *)RTA_DATA(rta) = ifi_ns; } - -next: - if (nh->nlmsg_type == NLMSG_DONE) - break; } - if (op == NL_DUP) { - char resp[NLBUFSIZ]; - unsigned i; - - nh = (struct nlmsghdr *)buf; - /* Routes might have dependencies between each other, and the - * kernel processes RTM_NEWROUTE messages sequentially. For n - * valid routes, we might need to send up to n requests to get - * all of them inserted. Routes that have been already inserted - * won't cause the whole request to fail, so we can simply - * repeat the whole request. This approach avoids the need to - * calculate dependencies: let the kernel do that. - */ - for (i = 0; i < dup_routes; i++) - nl_req(1, resp, nh, nlmsgs_size); - } + nh = (struct nlmsghdr *)buf; + /* Routes might have dependencies between each other, and the + * kernel processes RTM_NEWROUTE messages sequentially. For n + * valid routes, we might need to send up to n requests to get + * all of them inserted. Routes that have been already + * inserted won't cause the whole request to fail, so we can + * simply repeat the whole request. This approach avoids the + * need to calculate dependencies: let the kernel do that. + */ + for (i = 0; i < dup_routes; i++) + nl_req(1, resp, nh, nlmsgs_size); } /** diff --git a/netlink.h b/netlink.h index 5ac972d..36bbf9f 100644 --- a/netlink.h +++ b/netlink.h @@ -6,16 +6,11 @@ #ifndef NETLINK_H #define NETLINK_H -enum nl_op { - NL_GET, - NL_SET, - NL_DUP, -}; - void nl_sock_init(const struct ctx *c, bool ns); unsigned int nl_get_ext_if(sa_family_t af); -void nl_route(enum nl_op op, unsigned int ifi, unsigned int ifi_ns, - sa_family_t af, void *gw); +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); diff --git a/pasta.c b/pasta.c index 1a8f09c..14ecc71 100644 --- a/pasta.c +++ b/pasta.c @@ -281,8 +281,6 @@ void pasta_ns_conf(struct ctx *c) nl_link_set_mac(1, c->pasta_ifi, c->mac_guest); if (c->pasta_conf_ns) { - enum nl_op op_routes = c->no_copy_routes ? NL_SET : NL_DUP; - nl_link_up(1, c->pasta_ifi, c->mtu); if (c->ifi4) { @@ -292,8 +290,11 @@ void pasta_ns_conf(struct ctx *c) else nl_addr_dup(c->ifi4, c->pasta_ifi, AF_INET); - nl_route(op_routes, c->ifi4, c->pasta_ifi, AF_INET, - &c->ip4.gw); + if (c->no_copy_routes) + nl_route_set_def(c->pasta_ifi, AF_INET, + &c->ip4.gw); + else + nl_route_dup(c->ifi4, c->pasta_ifi, AF_INET); } if (c->ifi6) { @@ -302,8 +303,11 @@ void pasta_ns_conf(struct ctx *c) else nl_addr_dup(c->ifi4, c->pasta_ifi, AF_INET6); - nl_route(op_routes, c->ifi6, c->pasta_ifi, AF_INET6, - &c->ip6.gw); + if (c->no_copy_routes) + nl_route_set_def(c->pasta_ifi, AF_INET6, + &c->ip6.gw); + else + nl_route_dup(c->ifi6, c->pasta_ifi, AF_INET6); } } -- 2.41.0