public inbox for passt-dev@passt.top
 help / color / mirror / code / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: Stefano Brivio <sbrivio@redhat.com>, passt-dev@passt.top
Cc: David Gibson <david@gibson.dropbear.id.au>
Subject: [PATCH 03/17] netlink: Split nl_route() into separate operation functions
Date: Mon, 24 Jul 2023 16:09:22 +1000	[thread overview]
Message-ID: <20230724060936.952659-4-david@gibson.dropbear.id.au> (raw)
In-Reply-To: <20230724060936.952659-1-david@gibson.dropbear.id.au>

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 <david@gibson.dropbear.id.au>
---
 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);
 		}
 	}
 
-- 
@@ -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


  parent reply	other threads:[~2023-07-24  6:09 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-24  6:09 [PATCH 00/17] netlink fixes and cleanups David Gibson
2023-07-24  6:09 ` [PATCH 01/17] netlink: Split up functionality if nl_link() David Gibson
2023-08-02 22:47   ` Stefano Brivio
2023-08-03  2:09     ` David Gibson
2023-08-03  4:29       ` David Gibson
2023-08-03  5:39         ` David Gibson
2023-08-03  5:40         ` Stefano Brivio
2023-07-24  6:09 ` [PATCH 02/17] netlink: Split nl_addr() into separate operation functions David Gibson
2023-08-02 22:47   ` Stefano Brivio
2023-08-03  2:11     ` David Gibson
2023-07-24  6:09 ` David Gibson [this message]
2023-08-02 22:47   ` [PATCH 03/17] netlink: Split nl_route() " Stefano Brivio
2023-08-03  2:18     ` David Gibson
2023-07-24  6:09 ` [PATCH 04/17] netlink: Use struct in_addr for IPv4 addresses, not bare uint32_t David Gibson
2023-07-24  6:09 ` [PATCH 05/17] netlink: Explicitly pass netlink sockets to operations David Gibson
2023-07-24  6:09 ` [PATCH 06/17] netlink: Make nl_*_dup() use a separate datagram for each request David Gibson
2023-07-24  6:09 ` [PATCH 07/17] netlink: Start sequence number from 1 instead of 0 David Gibson
2023-07-24  6:09 ` [PATCH 08/17] netlink: Treat send() or recv() errors as fatal David Gibson
2023-08-02 22:47   ` Stefano Brivio
2023-08-03  2:19     ` David Gibson
2023-07-24  6:09 ` [PATCH 09/17] netlink: Fill in netlink header fields from nl_req() David Gibson
2023-07-24  6:09 ` [PATCH 10/17] netlink: Add nl_do() helper for simple operations with error checking David Gibson
2023-08-02 22:48   ` Stefano Brivio
2023-08-03  2:24     ` David Gibson
2023-07-24  6:09 ` [PATCH 11/17] netlink: Clearer reasoning about the netlink response buffer size David Gibson
2023-08-02 22:48   ` Stefano Brivio
2023-08-03  2:22     ` David Gibson
2023-07-24  6:09 ` [PATCH 12/17] netlink: Split nl_req() to allow processing multiple response datagrams David Gibson
2023-07-24  6:09 ` [PATCH 13/17] netlink: Add nl_foreach_oftype to filter response message types David Gibson
2023-07-24  6:09 ` [PATCH 14/17] netlink: Propagate errors for "set" operations David Gibson
2023-07-24  6:09 ` [PATCH 15/17] netlink: Always process all responses to a netlink request David Gibson
2023-07-24  6:09 ` [PATCH 16/17] netlink: Propagate errors for "dump" operations David Gibson
2023-07-24  6:09 ` [PATCH 17/17] netlink: Propagate errors for "dup" operations David Gibson
2023-08-02 22:48   ` Stefano Brivio
2023-08-03  2:26     ` David Gibson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230724060936.952659-4-david@gibson.dropbear.id.au \
    --to=david@gibson.dropbear.id.au \
    --cc=passt-dev@passt.top \
    --cc=sbrivio@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://passt.top/passt

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).