Revert "xfrm: xfrm_state_mtu should return at least 1280 for ipv6"

commit a6d95c5a628a09be129f25d5663a7e9db8261f51 upstream.

This reverts commit b515d2637276a3810d6595e10ab02c13bfd0b63a.

Commit b515d2637276a3810d6595e10ab02c13bfd0b63a ("xfrm: xfrm_state_mtu
should return at least 1280 for ipv6") in v5.14 breaks the TCP MSS
calculation in ipsec transport mode, resulting complete stalls of TCP
connections. This happens when the (P)MTU is 1280 or slighly larger.

The desired formula for the MSS is:
MSS = (MTU - ESP_overhead) - IP header - TCP header

However, the above commit clamps the (MTU - ESP_overhead) to a
minimum of 1280, turning the formula into
MSS = max(MTU - ESP overhead, 1280) -  IP header - TCP header

With the (P)MTU near 1280, the calculated MSS is too large and the
resulting TCP packets never make it to the destination because they
are over the actual PMTU.

The above commit also causes suboptimal double fragmentation in
xfrm tunnel mode, as described in
https://lore.kernel.org/netdev/20210429202529.codhwpc7w6kbudug@dwarf.suse.cz/

The original problem the above commit was trying to fix is now fixed
by commit 6596a0229541270fb8d38d989f91b78838e5e9da ("xfrm: fix MTU
regression").

Signed-off-by: Jiri Bohac <jbohac@suse.cz>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

4 files changed, 4 insertions, 15 deletions

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 337d29875e51..4a2843441caf 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1551,7 +1551,6 @@ void xfrm_sad_getinfo(struct net *net, struct xfrmk_sadinfo *si);
 void xfrm_spd_getinfo(struct net *net, struct xfrmk_spdinfo *si);
 u32 xfrm_replay_seqhi(struct xfrm_state *x, __be32 net_seq);
 int xfrm_init_replay(struct xfrm_state *x);
-u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu);
 u32 xfrm_state_mtu(struct xfrm_state *x, int mtu);
 int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload);
 int xfrm_init_state(struct xfrm_state *x);
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index ed9857b2875d..4b834bbf95e0 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -673,7 +673,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 		struct xfrm_dst *dst = (struct xfrm_dst *)skb_dst(skb);
 		u32 padto;
 
-		padto = min(x->tfcpad, __xfrm_state_mtu(x, dst->child_mtu_cached));
+		padto = min(x->tfcpad, xfrm_state_mtu(x, dst->child_mtu_cached));
 		if (skb->len < padto)
 			esp.tfclen = padto - skb->len;
 	}
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 7f2ffc7b1f75..fc8acb15dcfb 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -708,7 +708,7 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb)
 		struct xfrm_dst *dst = (struct xfrm_dst *)skb_dst(skb);
 		u32 padto;
 
-		padto = min(x->tfcpad, __xfrm_state_mtu(x, dst->child_mtu_cached));
+		padto = min(x->tfcpad, xfrm_state_mtu(x, dst->child_mtu_cached));
 		if (skb->len < padto)
 			esp.tfclen = padto - skb->len;
 	}
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 65e2805fa113..f5b846a2edcd 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -2537,7 +2537,7 @@ void xfrm_state_delete_tunnel(struct xfrm_state *x)
 }
 EXPORT_SYMBOL(xfrm_state_delete_tunnel);
 
-u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu)
+u32 xfrm_state_mtu(struct xfrm_state *x, int mtu)
 {
 	const struct xfrm_type *type = READ_ONCE(x->type);
 	struct crypto_aead *aead;
@@ -2568,17 +2568,7 @@ u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu)
 	return ((mtu - x->props.header_len - crypto_aead_authsize(aead) -
 		 net_adj) & ~(blksize - 1)) + net_adj - 2;
 }
-EXPORT_SYMBOL_GPL(__xfrm_state_mtu);
-
-u32 xfrm_state_mtu(struct xfrm_state *x, int mtu)
-{
-	mtu = __xfrm_state_mtu(x, mtu);
-
-	if (x->props.family == AF_INET6 && mtu < IPV6_MIN_MTU)
-		return IPV6_MIN_MTU;
-
-	return mtu;
-}
+EXPORT_SYMBOL_GPL(xfrm_state_mtu);
 
 int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload)
 {