resolv: fix gethostbyname2_r to match gethostbyname_r, fixing bugs with AAAA lookups

The latter half of gethostbyname2_r (doing AAAA queries) is rather dramatically different
from the corresponding portion of gethostbyname_r (doing A queries). This leads to problems
like calls to getaddrinfo only returning one IPv6 address, even when multiple exist.

Seems to be entirely a case of divergent evolution -- a half-decade of fixes for the IPv4
code but no love for IPv6. Until now. ;)

DNS behaviour for IPv6 is really no different than for IPv4 -- beyond the difference in
address sizes, there's no need for the functions to be so different.

Consequently, this patch really is almost just a cut-and-paste of gethostbyname_r, with
the appropriate substitutions of in6_addr, AF_INET6, etc; while holding on to the few
extra bits that actually belong in there (eg #ifdef __UCLIBC_HAS_IPV6__).

Signed-off-by: Wes Campaigne <westacular@gmail.com>

1 files changed, 104 insertions, 59 deletions

diff --git a/libc/inet/resolv.c b/libc/inet/resolv.c
index de038e400..3851b2a64 100644
--- a/libc/inet/resolv.c
+++ b/libc/inet/resolv.c
@@ -2200,12 +2200,13 @@ int gethostbyname2_r(const char *name,
 		? gethostbyname_r(name, result_buf, buf, buflen, result, h_errnop)
 		: HOST_NOT_FOUND;
 #else
-	struct in6_addr *in;
 	struct in6_addr **addr_list;
+	char **alias;
+	char *alias0;
 	unsigned char *packet;
 	struct resolv_answer a;
 	int i;
-	int nest = 0;
+	int packet_len;
 	int wrong_af = 0;
 
 	if (family == AF_INET)
@@ -2220,9 +2221,8 @@ int gethostbyname2_r(const char *name,
 
 	/* do /etc/hosts first */
 	{
-		int old_errno = errno;	/* Save the old errno and reset errno */
-		__set_errno(0);			/* to check for missing /etc/hosts. */
-
+		int old_errno = errno;  /* save the old errno and reset errno */
+		__set_errno(0);         /* to check for missing /etc/hosts. */
 		i = __get_hosts_byname_r(name, AF_INET6 /*family*/, result_buf,
 				buf, buflen, result, h_errnop);
 		if (i == NETDB_SUCCESS) {
@@ -2244,42 +2244,58 @@ int gethostbyname2_r(const char *name,
 		}
 		__set_errno(old_errno);
 	}
+
 	DPRINTF("Nothing found in /etc/hosts\n");
 
 	*h_errnop = NETDB_INTERNAL;
 
+	/* prepare future h_aliases[0] */
+	i = strlen(name) + 1;
+	if ((ssize_t)buflen <= i)
+		return ERANGE;
+	memcpy(buf, name, i); /* paranoia: name might change */
+	alias0 = buf;
+	buf += i;
+	buflen -= i;
 	/* make sure pointer is aligned */
 	i = ALIGN_BUFFER_OFFSET(buf);
 	buf += i;
 	buflen -= i;
 	/* Layout in buf:
-	 * struct in6_addr* in;
-	 * struct in6_addr* addr_list[2];
-	 * char scratch_buf[256];
+	 * char *alias[2];
+	 * struct in6_addr* addr_list[NN+1];
+	 * struct in6_addr* in[NN];
 	 */
-	in = (struct in6_addr*)buf;
-	buf += sizeof(*in);
-	buflen -= sizeof(*in);
-	addr_list = (struct in6_addr**)buf;
-	buf += sizeof(*addr_list) * 2;
-	buflen -= sizeof(*addr_list) * 2;
+	alias = (char **)buf;
+	buf += sizeof(alias[0]) * 2;
+	buflen -= sizeof(alias[0]) * 2;
+	addr_list = (struct in6_addr **)buf;
+	/* buflen may be < 0, must do signed compare */
 	if ((ssize_t)buflen < 256)
 		return ERANGE;
-	addr_list[0] = in;
-	addr_list[1] = NULL;
-	strncpy(buf, name, buflen);
-	buf[buflen] = '\0';
+
+	/* we store only one "alias" - the name itself */
+#ifdef __UCLIBC_MJN3_ONLY__
+#warning TODO -- generate the full list
+#endif
+	alias[0] = alias0;
+	alias[1] = NULL;
 
 	/* maybe it is already an address? */
-	if (inet_pton(AF_INET6, name, in)) {
-		result_buf->h_name = buf;
-		result_buf->h_addrtype = AF_INET6;
-		result_buf->h_length = sizeof(*in);
-		result_buf->h_addr_list = (char **) addr_list;
-		/* result_buf->h_aliases = ??? */
-		*result = result_buf;
-		*h_errnop = NETDB_SUCCESS;
-		return NETDB_SUCCESS;
+	{
+		struct in6_addr *in = (struct in6_addr *)(buf + sizeof(addr_list[0]) * 2);
+		if (inet_pton(AF_INET6, name, in)) {
+			addr_list[0] = in;
+			addr_list[1] = NULL;
+			result_buf->h_name = alias0;
+			result_buf->h_aliases = alias;
+			result_buf->h_addrtype = AF_INET6;
+			result_buf->h_length = sizeof(struct in6_addr);
+			result_buf->h_addr_list = (char **) addr_list;
+			*result = result_buf;
+			*h_errnop = NETDB_SUCCESS;
+			return NETDB_SUCCESS;
+		}
 	}
 
 	/* what if /etc/hosts has it but it's not IPv6?
@@ -2291,51 +2307,80 @@ int gethostbyname2_r(const char *name,
 	}
 
 	/* talk to DNS servers */
-/* TODO: why it's so different from gethostbyname_r (IPv4 case)? */
-	memset(&a, '\0', sizeof(a));
-	for (;;) {
-		int packet_len;
+	a.buf = buf;
+	/* take into account that at least one address will be there,
+	 * we'll need space of one in6_addr + two addr_list[] elems */
+	a.buflen = buflen - ((sizeof(addr_list[0]) * 2 + sizeof(struct in6_addr)));
+	a.add_count = 0;
+	packet_len = __dns_lookup(name, T_AAAA, &packet, &a);
+	if (packet_len < 0) {
+		*h_errnop = HOST_NOT_FOUND;
+		DPRINTF("__dns_lookup returned < 0\n");
+		return TRY_AGAIN;
+	}
 
-/* Hmm why we memset(a) to zeros only once? */
-		packet_len = __dns_lookup(buf, T_AAAA, &packet, &a);
-		if (packet_len < 0) {
-			*h_errnop = HOST_NOT_FOUND;
-			return TRY_AGAIN;
+	if (a.atype == T_AAAA) { /* ADDRESS */
+		/* we need space for addr_list[] and one IPv6 address */
+		/* + 1 accounting for 1st addr (it's in a.rdata),
+		 * another + 1 for NULL in last addr_list[]: */
+		int need_bytes = sizeof(addr_list[0]) * (a.add_count + 1 + 1)
+				/* for 1st addr (it's in a.rdata): */
+				+ sizeof(struct in6_addr);
+		/* how many bytes will 2nd and following addresses take? */
+		int ips_len = a.add_count * a.rdlength;
+
+		buflen -= (need_bytes + ips_len);
+		if ((ssize_t)buflen < 0) {
+			DPRINTF("buffer too small for all addresses\n");
+			/* *h_errnop = NETDB_INTERNAL; - already is */
+			i = ERANGE;
+			goto free_and_ret;
 		}
-		strncpy(buf, a.dotted, buflen);
-		free(a.dotted);
 
-		if (a.atype != T_CNAME)
-			break;
+		/* if there are additional addresses in buf,
+		 * move them forward so that they are not destroyed */
+		DPRINTF("a.add_count:%d a.rdlength:%d a.rdata:%p\n", a.add_count, a.rdlength, a.rdata);
+		memmove(buf + need_bytes, buf, ips_len);
 
-		DPRINTF("Got a CNAME in gethostbyname()\n");
-		if (++nest > MAX_RECURSE) {
-			*h_errnop = NO_RECOVERY;
-			return -1;
+		/* 1st address is in a.rdata, insert it  */
+		buf += need_bytes - sizeof(struct in6_addr);
+		memcpy(buf, a.rdata, sizeof(struct in6_addr));
+
+		/* fill addr_list[] */
+		for (i = 0; i <= a.add_count; i++) {
+			addr_list[i] = (struct in6_addr*)buf;
+			buf += sizeof(struct in6_addr);
 		}
-		i = __decode_dotted(packet, a.rdoffset, packet_len, buf, buflen);
-		free(packet);
-		if (i < 0) {
-			*h_errnop = NO_RECOVERY;
-			return -1;
+		addr_list[i] = NULL;
+
+		/* if we have enough space, we can report "better" name
+		 * (it may contain search domains attached by __dns_lookup,
+		 * or CNAME of the host if it is different from the name
+		 * we used to find it) */
+		if (a.dotted && buflen > strlen(a.dotted)) {
+			strcpy(buf, a.dotted);
+			alias0 = buf;
 		}
-	}
-	if (a.atype == T_AAAA) {	/* ADDRESS */
-		memcpy(in, a.rdata, sizeof(*in));
-		result_buf->h_name = buf;
+
+		result_buf->h_name = alias0;
+		result_buf->h_aliases = alias;
 		result_buf->h_addrtype = AF_INET6;
-		result_buf->h_length = sizeof(*in);
+		result_buf->h_length = sizeof(struct in6_addr);
 		result_buf->h_addr_list = (char **) addr_list;
-		/* result_buf->h_aliases = ??? */
-		free(packet);
 		*result = result_buf;
 		*h_errnop = NETDB_SUCCESS;
-		return NETDB_SUCCESS;
+		i = NETDB_SUCCESS;
+		goto free_and_ret;
 	}
-	free(packet);
+
 	*h_errnop = HOST_NOT_FOUND;
-	return TRY_AGAIN;
+	__set_h_errno(HOST_NOT_FOUND);
+	i = TRY_AGAIN;
 
+ free_and_ret:
+	free(a.dotted);
+	free(packet);
+	return i;
 #endif /* __UCLIBC_HAS_IPV6__ */
 }
 libc_hidden_def(gethostbyname2_r)