ip_tunnel: convert __be16 tunnel flags to bitmaps

Historically, tunnel flags like TUNNEL_CSUM or TUNNEL_ERSPAN_OPT
have been defined as __be16. Now all of those 16 bits are occupied
and there's no more free space for new flags.
It can't be simply switched to a bigger container with no
adjustments to the values, since it's an explicit Endian storage,
and on LE systems (__be16)0x0001 equals to
(__be64)0x0001000000000000.
We could probably define new 64-bit flags depending on the
Endianness, i.e. (__be64)0x0001 on BE and (__be64)0x00010000... on
LE, but that would introduce an Endianness dependency and spawn a
ton of Sparse warnings. To mitigate them, all of those places which
were adjusted with this change would be touched anyway, so why not
define stuff properly if there's no choice.

Define IP_TUNNEL_*_BIT counterparts as a bit number instead of the
value already coded and a fistful of <16 <-> bitmap> converters and
helpers. The two flags which have a different bit position are
SIT_ISATAP_BIT and VTI_ISVTI_BIT, as they were defined not as
__cpu_to_be16(), but as (__force __be16), i.e. had different
positions on LE and BE. Now they both have strongly defined places.
Change all __be16 fields which were used to store those flags, to
IP_TUNNEL_DECLARE_FLAGS() -> DECLARE_BITMAP(__IP_TUNNEL_FLAG_NUM) ->
unsigned long[1] for now, and replace all TUNNEL_* occurrences to
their bitmap counterparts. Use the converters in the places which talk
to the userspace, hardware (NFP) or other hosts (GRE header). The rest
must explicitly use the new flags only. This must be done at once,
otherwise there will be too many conversions throughout the code in
the intermediate commits.
Finally, disable the old __be16 flags for use in the kernel code
(except for the two 'irregular' flags mentioned above), to prevent
any accidental (mis)use of them. For the userspace, nothing is
changed, only additions were made.

Most noticeable bloat-o-meter difference (.text):

vmlinux:	307/-1 (306)
gre.ko:		62/0 (62)
ip_gre.ko:	941/-217 (724)	[*]
ip_tunnel.ko:	390/-900 (-510)	[**]
ip_vti.ko:	138/0 (138)
ip6_gre.ko:	534/-18 (516)	[*]
ip6_tunnel.ko:	118/-10 (108)

[*] gre_flags_to_tnl_flags() grew, but still is inlined
[**] ip_tunnel_find() got uninlined, hence such decrease

The average code size increase in non-extreme case is 100-200 bytes
per module, mostly due to sizeof(long) > sizeof(__be16), as
%__IP_TUNNEL_FLAG_NUM is less than %BITS_PER_LONG and the compilers
are able to expand the majority of bitmap_*() calls here into direct
operations on scalars.

Reviewed-by: Simon Horman <[email protected]>
Signed-off-by: Alexander Lobakin <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

drivers/net/bareudp.c

@@ -61,6 +61,7 @@ struct bareudp_dev {
 static int bareudp_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct metadata_dst *tun_dst = NULL;
+	IP_TUNNEL_DECLARE_FLAGS(key) = { };
 	struct bareudp_dev *bareudp;
 	unsigned short family;
 	unsigned int len;
@@ -137,7 +138,10 @@ static int bareudp_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		bareudp->dev->stats.rx_dropped++;
 		goto drop;
 	}
-	tun_dst = udp_tun_rx_dst(skb, family, TUNNEL_KEY, 0, 0);
+
+	__set_bit(IP_TUNNEL_KEY_BIT, key);
+
+	tun_dst = udp_tun_rx_dst(skb, family, key, 0, 0);
 	if (!tun_dst) {
 		bareudp->dev->stats.rx_dropped++;
 		goto drop;
@@ -285,10 +289,10 @@ static int bareudp_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 			    struct bareudp_dev *bareudp,
 			    const struct ip_tunnel_info *info)
 {
+	bool udp_sum = test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags);
 	bool xnet = !net_eq(bareudp->net, dev_net(bareudp->dev));
 	bool use_cache = ip_tunnel_dst_cache_usable(skb, info);
 	struct socket *sock = rcu_dereference(bareudp->sock);
-	bool udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM);
 	const struct ip_tunnel_key *key = &info->key;
 	struct rtable *rt;
 	__be16 sport, df;
@@ -316,7 +320,8 @@ static int bareudp_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 
 	tos = ip_tunnel_ecn_encap(key->tos, ip_hdr(skb), skb);
 	ttl = key->ttl;
-	df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0;
+	df = test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, key->tun_flags) ?
+	     htons(IP_DF) : 0;
 	skb_scrub_packet(skb, xnet);
 
 	err = -ENOSPC;
@@ -338,7 +343,8 @@ static int bareudp_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 	udp_tunnel_xmit_skb(rt, sock->sk, skb, saddr, info->key.u.ipv4.dst,
 			    tos, ttl, df, sport, bareudp->port,
 			    !net_eq(bareudp->net, dev_net(bareudp->dev)),
-			    !(info->key.tun_flags & TUNNEL_CSUM));
+			    !test_bit(IP_TUNNEL_CSUM_BIT,
+				      info->key.tun_flags));
 	return 0;
 
 free_dst:
@@ -350,10 +356,10 @@ static int bareudp6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 			     struct bareudp_dev *bareudp,
 			     const struct ip_tunnel_info *info)
 {
+	bool udp_sum = test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags);
 	bool xnet = !net_eq(bareudp->net, dev_net(bareudp->dev));
 	bool use_cache = ip_tunnel_dst_cache_usable(skb, info);
 	struct socket *sock  = rcu_dereference(bareudp->sock);
-	bool udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM);
 	const struct ip_tunnel_key *key = &info->key;
 	struct dst_entry *dst = NULL;
 	struct in6_addr saddr, daddr;
@@ -402,7 +408,8 @@ static int bareudp6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 	udp_tunnel6_xmit_skb(dst, sock->sk, skb, dev,
 			     &saddr, &daddr, prio, ttl,
 			     info->key.label, sport, bareudp->port,
-			     !(info->key.tun_flags & TUNNEL_CSUM));
+			     !test_bit(IP_TUNNEL_CSUM_BIT,
+				       info->key.tun_flags));
 	return 0;
 
 free_dst:

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h

@@ -117,7 +117,7 @@ bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a,
 
 bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a,
 					   struct mlx5e_encap_key *b,
-					   __be16 tun_flags);
+					   u32 tun_type);
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif //__MLX5_EN_TC_TUNNEL_H__

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c

@@ -587,7 +587,7 @@ bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a,
 
 bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a,
 					   struct mlx5e_encap_key *b,
-					   __be16 tun_flags)
+					   u32 tun_type)
 {
 	struct ip_tunnel_info *a_info;
 	struct ip_tunnel_info *b_info;
@@ -596,8 +596,8 @@ bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a,
 	if (!mlx5e_tc_tun_encap_info_equal_generic(a, b))
 		return false;
 
-	a_has_opts = !!(a->ip_tun_key->tun_flags & tun_flags);
-	b_has_opts = !!(b->ip_tun_key->tun_flags & tun_flags);
+	a_has_opts = test_bit(tun_type, a->ip_tun_key->tun_flags);
+	b_has_opts = test_bit(tun_type, b->ip_tun_key->tun_flags);
 
 	/* keys are equal when both don't have any options attached */
 	if (!a_has_opts && !b_has_opts)

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c

@@ -106,12 +106,13 @@ static int mlx5e_gen_ip_tunnel_header_geneve(char buf[],
 	memset(geneveh, 0, sizeof(*geneveh));
 	geneveh->ver = MLX5E_GENEVE_VER;
 	geneveh->opt_len = tun_info->options_len / 4;
-	geneveh->oam = !!(tun_info->key.tun_flags & TUNNEL_OAM);
-	geneveh->critical = !!(tun_info->key.tun_flags & TUNNEL_CRIT_OPT);
+	geneveh->oam = test_bit(IP_TUNNEL_OAM_BIT, tun_info->key.tun_flags);
+	geneveh->critical = test_bit(IP_TUNNEL_CRIT_OPT_BIT,
+				     tun_info->key.tun_flags);
 	mlx5e_tunnel_id_to_vni(tun_info->key.tun_id, geneveh->vni);
 	geneveh->proto_type = htons(ETH_P_TEB);
 
-	if (tun_info->key.tun_flags & TUNNEL_GENEVE_OPT) {
+	if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, tun_info->key.tun_flags)) {
 		if (!geneveh->opt_len)
 			return -EOPNOTSUPP;
 		ip_tunnel_info_opts_get(geneveh->options, tun_info);
@@ -188,7 +189,7 @@ static int mlx5e_tc_tun_parse_geneve_options(struct mlx5e_priv *priv,
 
 	/* make sure that we're talking about GENEVE options */
 
-	if (enc_opts.key->dst_opt_type != TUNNEL_GENEVE_OPT) {
+	if (enc_opts.key->dst_opt_type != IP_TUNNEL_GENEVE_OPT_BIT) {
 		NL_SET_ERR_MSG_MOD(extack,
 				   "Matching on GENEVE options: option type is not GENEVE");
 		netdev_warn(priv->netdev,
@@ -337,7 +338,8 @@ static int mlx5e_tc_tun_parse_geneve(struct mlx5e_priv *priv,
 static bool mlx5e_tc_tun_encap_info_equal_geneve(struct mlx5e_encap_key *a,
 						 struct mlx5e_encap_key *b)
 {
-	return mlx5e_tc_tun_encap_info_equal_options(a, b, TUNNEL_GENEVE_OPT);
+	return mlx5e_tc_tun_encap_info_equal_options(a, b,
+						     IP_TUNNEL_GENEVE_OPT_BIT);
 }
 
 struct mlx5e_tc_tunnel geneve_tunnel = {

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c

@@ -31,20 +31,24 @@ static int mlx5e_gen_ip_tunnel_header_gretap(char buf[],
 	const struct ip_tunnel_key *tun_key  = &e->tun_info->key;
 	struct gre_base_hdr *greh = (struct gre_base_hdr *)(buf);
 	__be32 tun_id = tunnel_id_to_key32(tun_key->tun_id);
+	IP_TUNNEL_DECLARE_FLAGS(unsupp) = { };
 	int hdr_len;
 
 	*ip_proto = IPPROTO_GRE;
 
 	/* the HW does not calculate GRE csum or sequences */
-	if (tun_key->tun_flags & (TUNNEL_CSUM | TUNNEL_SEQ))
+	__set_bit(IP_TUNNEL_CSUM_BIT, unsupp);
+	__set_bit(IP_TUNNEL_SEQ_BIT, unsupp);
+
+	if (ip_tunnel_flags_intersect(tun_key->tun_flags, unsupp))
 		return -EOPNOTSUPP;
 
 	greh->protocol = htons(ETH_P_TEB);
 
 	/* GRE key */
 	hdr_len	= mlx5e_tc_tun_calc_hlen_gretap(e);
 	greh->flags = gre_tnl_flags_to_gre_flags(tun_key->tun_flags);
-	if (tun_key->tun_flags & TUNNEL_KEY) {
+	if (test_bit(IP_TUNNEL_KEY_BIT, tun_key->tun_flags)) {
 		__be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4);
 		*ptr = tun_id;
 	}

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c

@@ -90,7 +90,7 @@ static int mlx5e_gen_ip_tunnel_header_vxlan(char buf[],
 	const struct vxlan_metadata *md;
 	struct vxlanhdr *vxh;
 
-	if ((tun_key->tun_flags & TUNNEL_VXLAN_OPT) &&
+	if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, tun_key->tun_flags) &&
 	    e->tun_info->options_len != sizeof(*md))
 		return -EOPNOTSUPP;
 	vxh = (struct vxlanhdr *)((char *)udp + sizeof(struct udphdr));
@@ -99,7 +99,7 @@ static int mlx5e_gen_ip_tunnel_header_vxlan(char buf[],
 	udp->dest = tun_key->tp_dst;
 	vxh->vx_flags = VXLAN_HF_VNI;
 	vxh->vx_vni = vxlan_vni_field(tun_id);
-	if (tun_key->tun_flags & TUNNEL_VXLAN_OPT) {
+	if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, tun_key->tun_flags)) {
 		md = ip_tunnel_info_opts(e->tun_info);
 		vxlan_build_gbp_hdr(vxh, md);
 	}
@@ -125,7 +125,7 @@ static int mlx5e_tc_tun_parse_vxlan_gbp_option(struct mlx5e_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
-	if (enc_opts.key->dst_opt_type != TUNNEL_VXLAN_OPT) {
+	if (enc_opts.key->dst_opt_type != IP_TUNNEL_VXLAN_OPT_BIT) {
 		NL_SET_ERR_MSG_MOD(extack, "Wrong VxLAN option type: not GBP");
 		return -EOPNOTSUPP;
 	}
@@ -208,7 +208,8 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv,
 static bool mlx5e_tc_tun_encap_info_equal_vxlan(struct mlx5e_encap_key *a,
 						struct mlx5e_encap_key *b)
 {
-	return mlx5e_tc_tun_encap_info_equal_options(a, b, TUNNEL_VXLAN_OPT);
+	return mlx5e_tc_tun_encap_info_equal_options(a, b,
+						     IP_TUNNEL_VXLAN_OPT_BIT);
 }
 
 static int mlx5e_tc_tun_get_remote_ifindex(struct net_device *mirred_dev)

drivers/net/ethernet/mellanox/mlx5/core/en_tc.c

@@ -5464,13 +5464,16 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb
 	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct tunnel_match_enc_opts enc_opts = {};
 	struct mlx5_rep_uplink_priv *uplink_priv;
+	IP_TUNNEL_DECLARE_FLAGS(flags) = { };
 	struct mlx5e_rep_priv *uplink_rpriv;
 	struct metadata_dst *tun_dst;
 	struct tunnel_match_key key;
 	u32 tun_id, enc_opts_id;
 	struct net_device *dev;
 	int err;
 
+	__set_bit(IP_TUNNEL_KEY_BIT, flags);
+
 	enc_opts_id = tunnel_id & ENC_OPTS_BITS_MASK;
 	tun_id = tunnel_id >> ENC_OPTS_BITS;
 
@@ -5503,14 +5506,14 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb
 	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
 		tun_dst = __ip_tun_set_dst(key.enc_ipv4.src, key.enc_ipv4.dst,
 					   key.enc_ip.tos, key.enc_ip.ttl,
-					   key.enc_tp.dst, TUNNEL_KEY,
+					   key.enc_tp.dst, flags,
 					   key32_to_tunnel_id(key.enc_key_id.keyid),
 					   enc_opts.key.len);
 		break;
 	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
 		tun_dst = __ipv6_tun_set_dst(&key.enc_ipv6.src, &key.enc_ipv6.dst,
 					     key.enc_ip.tos, key.enc_ip.ttl,
-					     key.enc_tp.dst, 0, TUNNEL_KEY,
+					     key.enc_tp.dst, 0, flags,
 					     key32_to_tunnel_id(key.enc_key_id.keyid),
 					     enc_opts.key.len);
 		break;
@@ -5528,11 +5531,16 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb
 
 	tun_dst->u.tun_info.key.tp_src = key.enc_tp.src;
 
-	if (enc_opts.key.len)
+	if (enc_opts.key.len) {
+		ip_tunnel_flags_zero(flags);
+		if (enc_opts.key.dst_opt_type)
+			__set_bit(enc_opts.key.dst_opt_type, flags);
+
 		ip_tunnel_info_opts_set(&tun_dst->u.tun_info,
 					enc_opts.key.data,
 					enc_opts.key.len,
-					enc_opts.key.dst_opt_type);
+					flags);
+	}
 
 	skb_dst_set(skb, (struct dst_entry *)tun_dst);
 	dev = dev_get_by_index(&init_net, key.filter_ifindex);

drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c

@@ -27,23 +27,23 @@ mlxsw_sp_ipip_netdev_parms6(const struct net_device *ol_dev)
 static bool
 mlxsw_sp_ipip_parms4_has_ikey(const struct ip_tunnel_parm_kern *parms)
 {
-	return !!(parms->i_flags & TUNNEL_KEY);
+	return test_bit(IP_TUNNEL_KEY_BIT, parms->i_flags);
 }
 
 static bool mlxsw_sp_ipip_parms6_has_ikey(const struct __ip6_tnl_parm *parms)
 {
-	return !!(parms->i_flags & TUNNEL_KEY);
+	return test_bit(IP_TUNNEL_KEY_BIT, parms->i_flags);
 }
 
 static bool
 mlxsw_sp_ipip_parms4_has_okey(const struct ip_tunnel_parm_kern *parms)
 {
-	return !!(parms->o_flags & TUNNEL_KEY);
+	return test_bit(IP_TUNNEL_KEY_BIT, parms->o_flags);
 }
 
 static bool mlxsw_sp_ipip_parms6_has_okey(const struct __ip6_tnl_parm *parms)
 {
-	return !!(parms->o_flags & TUNNEL_KEY);
+	return test_bit(IP_TUNNEL_KEY_BIT, parms->o_flags);
 }
 
 static u32 mlxsw_sp_ipip_parms4_ikey(const struct ip_tunnel_parm_kern *parms)
@@ -242,12 +242,15 @@ static bool mlxsw_sp_ipip_can_offload_gre4(const struct mlxsw_sp *mlxsw_sp,
 					   const struct net_device *ol_dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(ol_dev);
-	__be16 okflags = TUNNEL_KEY; /* We can't offload any other features. */
 	bool inherit_ttl = tunnel->parms.iph.ttl == 0;
 	bool inherit_tos = tunnel->parms.iph.tos & 0x1;
+	IP_TUNNEL_DECLARE_FLAGS(okflags) = { };
 
-	return (tunnel->parms.i_flags & ~okflags) == 0 &&
-	       (tunnel->parms.o_flags & ~okflags) == 0 &&
+	/* We can't offload any other features. */
+	__set_bit(IP_TUNNEL_KEY_BIT, okflags);
+
+	return ip_tunnel_flags_subset(tunnel->parms.i_flags, okflags) &&
+	       ip_tunnel_flags_subset(tunnel->parms.o_flags, okflags) &&
 	       inherit_ttl && inherit_tos &&
 	       mlxsw_sp_ipip_tunnel_complete(MLXSW_SP_L3_PROTO_IPV4, ol_dev);
 }
@@ -443,10 +446,13 @@ static bool mlxsw_sp_ipip_can_offload_gre6(const struct mlxsw_sp *mlxsw_sp,
 	struct __ip6_tnl_parm tparm = mlxsw_sp_ipip_netdev_parms6(ol_dev);
 	bool inherit_tos = tparm.flags & IP6_TNL_F_USE_ORIG_TCLASS;
 	bool inherit_ttl = tparm.hop_limit == 0;
-	__be16 okflags = TUNNEL_KEY; /* We can't offload any other features. */
+	IP_TUNNEL_DECLARE_FLAGS(okflags) = { };
+
+	/* We can't offload any other features. */
+	__set_bit(IP_TUNNEL_KEY_BIT, okflags);
 
-	return (tparm.i_flags & ~okflags) == 0 &&
-	       (tparm.o_flags & ~okflags) == 0 &&
+	return ip_tunnel_flags_subset(tparm.i_flags, okflags) &&
+	       ip_tunnel_flags_subset(tparm.o_flags, okflags) &&
 	       inherit_ttl && inherit_tos &&
 	       mlxsw_sp_ipip_tunnel_complete(MLXSW_SP_L3_PROTO_IPV6, ol_dev);
 }

drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c

@@ -461,7 +461,8 @@ mlxsw_sp_span_entry_gretap4_parms(struct mlxsw_sp *mlxsw_sp,
 
 	if (!(to_dev->flags & IFF_UP) ||
 	    /* Reject tunnels with GRE keys, checksums, etc. */
-	    tparm.i_flags || tparm.o_flags ||
+	    !ip_tunnel_flags_empty(tparm.i_flags) ||
+	    !ip_tunnel_flags_empty(tparm.o_flags) ||
 	    /* Require a fixed TTL and a TOS copied from the mirrored packet. */
 	    inherit_ttl || !inherit_tos ||
 	    /* A destination address may not be "any". */
@@ -565,7 +566,8 @@ mlxsw_sp_span_entry_gretap6_parms(struct mlxsw_sp *mlxsw_sp,
 
 	if (!(to_dev->flags & IFF_UP) ||
 	    /* Reject tunnels with GRE keys, checksums, etc. */
-	    tparm.i_flags || tparm.o_flags ||
+	    !ip_tunnel_flags_empty(tparm.i_flags) ||
+	    !ip_tunnel_flags_empty(tparm.o_flags) ||
 	    /* Require a fixed TTL and a TOS copied from the mirrored packet. */
 	    inherit_ttl || !inherit_tos ||
 	    /* A destination address may not be "any". */