1 files changed, 162 insertions, 202 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c
index 0dd17514caae..6859f1c1a831 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c
@@ -34,78 +34,15 @@
 #include <crypto/aead.h>
 #include <net/xfrm.h>
 #include <net/esp.h>
-
-#include "en_accel/ipsec_rxtx.h"
-#include "en_accel/ipsec.h"
-#include "accel/accel.h"
+#include "ipsec.h"
+#include "ipsec_rxtx.h"
 #include "en.h"
 
 enum {
-	MLX5E_IPSEC_RX_SYNDROME_DECRYPTED = 0x11,
-	MLX5E_IPSEC_RX_SYNDROME_AUTH_FAILED = 0x12,
-	MLX5E_IPSEC_RX_SYNDROME_BAD_PROTO = 0x17,
-};
-
-struct mlx5e_ipsec_rx_metadata {
-	unsigned char   nexthdr;
-	__be32		sa_handle;
-} __packed;
-
-enum {
 	MLX5E_IPSEC_TX_SYNDROME_OFFLOAD = 0x8,
 	MLX5E_IPSEC_TX_SYNDROME_OFFLOAD_WITH_LSO_TCP = 0x9,
 };
 
-struct mlx5e_ipsec_tx_metadata {
-	__be16 mss_inv;         /* 1/MSS in 16bit fixed point, only for LSO */
-	__be16 seq;             /* LSBs of the first TCP seq, only for LSO */
-	u8     esp_next_proto;  /* Next protocol of ESP */
-} __packed;
-
-struct mlx5e_ipsec_metadata {
-	unsigned char syndrome;
-	union {
-		unsigned char raw[5];
-		/* from FPGA to host, on successful decrypt */
-		struct mlx5e_ipsec_rx_metadata rx;
-		/* from host to FPGA */
-		struct mlx5e_ipsec_tx_metadata tx;
-	} __packed content;
-	/* packet type ID field	*/
-	__be16 ethertype;
-} __packed;
-
-#define MAX_LSO_MSS 2048
-
-/* Pre-calculated (Q0.16) fixed-point inverse 1/x function */
-static __be16 mlx5e_ipsec_inverse_table[MAX_LSO_MSS];
-
-static inline __be16 mlx5e_ipsec_mss_inv(struct sk_buff *skb)
-{
-	return mlx5e_ipsec_inverse_table[skb_shinfo(skb)->gso_size];
-}
-
-static struct mlx5e_ipsec_metadata *mlx5e_ipsec_add_metadata(struct sk_buff *skb)
-{
-	struct mlx5e_ipsec_metadata *mdata;
-	struct ethhdr *eth;
-
-	if (unlikely(skb_cow_head(skb, sizeof(*mdata))))
-		return ERR_PTR(-ENOMEM);
-
-	eth = (struct ethhdr *)skb_push(skb, sizeof(*mdata));
-	skb->mac_header -= sizeof(*mdata);
-	mdata = (struct mlx5e_ipsec_metadata *)(eth + 1);
-
-	memmove(skb->data, skb->data + sizeof(*mdata),
-		2 * ETH_ALEN);
-
-	eth->h_proto = cpu_to_be16(MLX5E_METADATA_ETHER_TYPE);
-
-	memset(mdata->content.raw, 0, sizeof(mdata->content.raw));
-	return mdata;
-}
-
 static int mlx5e_ipsec_remove_trailer(struct sk_buff *skb, struct xfrm_state *x)
 {
 	unsigned int alen = crypto_aead_authsize(x->data);
@@ -136,33 +73,78 @@ static void mlx5e_ipsec_set_swp(struct sk_buff *skb,
 				struct mlx5_wqe_eth_seg *eseg, u8 mode,
 				struct xfrm_offload *xo)
 {
-	struct mlx5e_swp_spec swp_spec = {};
-
 	/* Tunnel Mode:
 	 * SWP:      OutL3       InL3  InL4
 	 * Pkt: MAC  IP     ESP  IP    L4
 	 *
 	 * Transport Mode:
-	 * SWP:      OutL3       InL4
-	 *           InL3
+	 * SWP:      OutL3       OutL4
 	 * Pkt: MAC  IP     ESP  L4
+	 *
+	 * Tunnel(VXLAN TCP/UDP) over Transport Mode
+	 * SWP:      OutL3                   InL3  InL4
+	 * Pkt: MAC  IP     ESP  UDP  VXLAN  IP    L4
 	 */
-	swp_spec.l3_proto = skb->protocol;
-	swp_spec.is_tun = mode == XFRM_MODE_TUNNEL;
-	if (swp_spec.is_tun) {
-		if (xo->proto == IPPROTO_IPV6) {
-			swp_spec.tun_l3_proto = htons(ETH_P_IPV6);
-			swp_spec.tun_l4_proto = inner_ipv6_hdr(skb)->nexthdr;
-		} else {
-			swp_spec.tun_l3_proto = htons(ETH_P_IP);
-			swp_spec.tun_l4_proto = inner_ip_hdr(skb)->protocol;
+
+	/* Shared settings */
+	eseg->swp_outer_l3_offset = skb_network_offset(skb) / 2;
+	if (skb->protocol == htons(ETH_P_IPV6))
+		eseg->swp_flags |= MLX5_ETH_WQE_SWP_OUTER_L3_IPV6;
+
+	/* Tunnel mode */
+	if (mode == XFRM_MODE_TUNNEL) {
+		eseg->swp_inner_l3_offset = skb_inner_network_offset(skb) / 2;
+		if (xo->proto == IPPROTO_IPV6)
+			eseg->swp_flags |= MLX5_ETH_WQE_SWP_INNER_L3_IPV6;
+
+		switch (xo->inner_ipproto) {
+		case IPPROTO_UDP:
+			eseg->swp_flags |= MLX5_ETH_WQE_SWP_INNER_L4_UDP;
+			fallthrough;
+		case IPPROTO_TCP:
+			/* IP | ESP | IP | [TCP | UDP] */
+			eseg->swp_inner_l4_offset = skb_inner_transport_offset(skb) / 2;
+			break;
+		default:
+			break;
+		}
+		return;
+	}
+
+	/* Transport mode */
+	if (mode != XFRM_MODE_TRANSPORT)
+		return;
+
+	if (!xo->inner_ipproto) {
+		switch (xo->proto) {
+		case IPPROTO_UDP:
+			eseg->swp_flags |= MLX5_ETH_WQE_SWP_OUTER_L4_UDP;
+			fallthrough;
+		case IPPROTO_TCP:
+			/* IP | ESP | TCP */
+			eseg->swp_outer_l4_offset = skb_inner_transport_offset(skb) / 2;
+			break;
+		default:
+			break;
 		}
 	} else {
-		swp_spec.tun_l3_proto = skb->protocol;
-		swp_spec.tun_l4_proto = xo->proto;
+		/* Tunnel(VXLAN TCP/UDP) over Transport Mode */
+		switch (xo->inner_ipproto) {
+		case IPPROTO_UDP:
+			eseg->swp_flags |= MLX5_ETH_WQE_SWP_INNER_L4_UDP;
+			fallthrough;
+		case IPPROTO_TCP:
+			eseg->swp_inner_l3_offset = skb_inner_network_offset(skb) / 2;
+			eseg->swp_inner_l4_offset =
+				(skb->csum_start + skb->head - skb->data) / 2;
+			if (inner_ip_hdr(skb)->version == 6)
+				eseg->swp_flags |= MLX5_ETH_WQE_SWP_INNER_L3_IPV6;
+			break;
+		default:
+			break;
+		}
 	}
 
-	mlx5e_set_eseg_swp(skb, eseg, &swp_spec);
 }
 
 void mlx5e_ipsec_set_iv_esn(struct sk_buff *skb, struct xfrm_state *x,
@@ -199,54 +181,90 @@ void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_state *x,
 	skb_store_bits(skb, iv_offset, &seqno, 8);
 }
 
-static void mlx5e_ipsec_set_metadata(struct sk_buff *skb,
-				     struct mlx5e_ipsec_metadata *mdata,
-				     struct xfrm_offload *xo)
+void mlx5e_ipsec_handle_tx_wqe(struct mlx5e_tx_wqe *wqe,
+			       struct mlx5e_accel_tx_ipsec_state *ipsec_st,
+			       struct mlx5_wqe_inline_seg *inlseg)
 {
-	struct ip_esp_hdr *esph;
-	struct tcphdr *tcph;
-
-	if (skb_is_gso(skb)) {
-		/* Add LSO metadata indication */
-		esph = ip_esp_hdr(skb);
-		tcph = inner_tcp_hdr(skb);
-		netdev_dbg(skb->dev, "   Offloading GSO packet outer L3 %u; L4 %u; Inner L3 %u; L4 %u\n",
-			   skb->network_header,
-			   skb->transport_header,
-			   skb->inner_network_header,
-			   skb->inner_transport_header);
-		netdev_dbg(skb->dev, "   Offloading GSO packet of len %u; mss %u; TCP sp %u dp %u seq 0x%x ESP seq 0x%x\n",
-			   skb->len, skb_shinfo(skb)->gso_size,
-			   ntohs(tcph->source), ntohs(tcph->dest),
-			   ntohl(tcph->seq), ntohl(esph->seq_no));
-		mdata->syndrome = MLX5E_IPSEC_TX_SYNDROME_OFFLOAD_WITH_LSO_TCP;
-		mdata->content.tx.mss_inv = mlx5e_ipsec_mss_inv(skb);
-		mdata->content.tx.seq = htons(ntohl(tcph->seq) & 0xFFFF);
-	} else {
-		mdata->syndrome = MLX5E_IPSEC_TX_SYNDROME_OFFLOAD;
-	}
-	mdata->content.tx.esp_next_proto = xo->proto;
+	inlseg->byte_count = cpu_to_be32(ipsec_st->tailen | MLX5_INLINE_SEG);
+	esp_output_fill_trailer((u8 *)inlseg->data, 0, ipsec_st->plen, ipsec_st->xo->proto);
+}
+
+static int mlx5e_ipsec_set_state(struct mlx5e_priv *priv,
+				 struct sk_buff *skb,
+				 struct xfrm_state *x,
+				 struct xfrm_offload *xo,
+				 struct mlx5e_accel_tx_ipsec_state *ipsec_st)
+{
+	unsigned int blksize, clen, alen, plen;
+	struct crypto_aead *aead;
+	unsigned int tailen;
+
+	ipsec_st->x = x;
+	ipsec_st->xo = xo;
+	aead = x->data;
+	alen = crypto_aead_authsize(aead);
+	blksize = ALIGN(crypto_aead_blocksize(aead), 4);
+	clen = ALIGN(skb->len + 2, blksize);
+	plen = max_t(u32, clen - skb->len, 4);
+	tailen = plen + alen;
+	ipsec_st->plen = plen;
+	ipsec_st->tailen = tailen;
 
-	netdev_dbg(skb->dev, "   TX metadata syndrome %u proto %u mss_inv %04x seq %04x\n",
-		   mdata->syndrome, mdata->content.tx.esp_next_proto,
-		   ntohs(mdata->content.tx.mss_inv),
-		   ntohs(mdata->content.tx.seq));
+	return 0;
 }
 
-struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev,
-					  struct mlx5e_tx_wqe *wqe,
-					  struct sk_buff *skb)
+void mlx5e_ipsec_tx_build_eseg(struct mlx5e_priv *priv, struct sk_buff *skb,
+			       struct mlx5_wqe_eth_seg *eseg)
+{
+	struct xfrm_offload *xo = xfrm_offload(skb);
+	struct xfrm_encap_tmpl  *encap;
+	struct xfrm_state *x;
+	struct sec_path *sp;
+	u8 l3_proto;
+
+	sp = skb_sec_path(skb);
+	if (unlikely(sp->len != 1))
+		return;
+
+	x = xfrm_input_state(skb);
+	if (unlikely(!x))
+		return;
+
+	if (unlikely(!x->xso.offload_handle ||
+		     (skb->protocol != htons(ETH_P_IP) &&
+		      skb->protocol != htons(ETH_P_IPV6))))
+		return;
+
+	mlx5e_ipsec_set_swp(skb, eseg, x->props.mode, xo);
+
+	l3_proto = (x->props.family == AF_INET) ?
+		   ((struct iphdr *)skb_network_header(skb))->protocol :
+		   ((struct ipv6hdr *)skb_network_header(skb))->nexthdr;
+
+	eseg->flow_table_metadata |= cpu_to_be32(MLX5_ETH_WQE_FT_META_IPSEC);
+	eseg->trailer |= cpu_to_be32(MLX5_ETH_WQE_INSERT_TRAILER);
+	encap = x->encap;
+	if (!encap) {
+		eseg->trailer |= (l3_proto == IPPROTO_ESP) ?
+			cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_OUTER_IP_ASSOC) :
+			cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_OUTER_L4_ASSOC);
+	} else if (encap->encap_type == UDP_ENCAP_ESPINUDP) {
+		eseg->trailer |= (l3_proto == IPPROTO_ESP) ?
+			cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_INNER_IP_ASSOC) :
+			cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_INNER_L4_ASSOC);
+	}
+}
+
+bool mlx5e_ipsec_handle_tx_skb(struct net_device *netdev,
+			       struct sk_buff *skb,
+			       struct mlx5e_accel_tx_ipsec_state *ipsec_st)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 	struct xfrm_offload *xo = xfrm_offload(skb);
-	struct mlx5e_ipsec_metadata *mdata;
 	struct mlx5e_ipsec_sa_entry *sa_entry;
 	struct xfrm_state *x;
 	struct sec_path *sp;
 
-	if (!xo)
-		return skb;
-
 	sp = skb_sec_path(skb);
 	if (unlikely(sp->len != 1)) {
 		atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_bundle);
@@ -271,124 +289,66 @@ struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev,
 			atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_trailer);
 			goto drop;
 		}
-	mdata = mlx5e_ipsec_add_metadata(skb);
-	if (IS_ERR(mdata)) {
-		atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_metadata);
-		goto drop;
-	}
-	mlx5e_ipsec_set_swp(skb, &wqe->eth, x->props.mode, xo);
+
 	sa_entry = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle;
 	sa_entry->set_iv_op(skb, x, xo);
-	mlx5e_ipsec_set_metadata(skb, mdata, xo);
+	mlx5e_ipsec_set_state(priv, skb, x, xo, ipsec_st);
 
-	return skb;
+	return true;
 
 drop:
 	kfree_skb(skb);
-	return NULL;
+	return false;
 }
 
-static inline struct xfrm_state *
-mlx5e_ipsec_build_sp(struct net_device *netdev, struct sk_buff *skb,
-		     struct mlx5e_ipsec_metadata *mdata)
+enum {
+	MLX5E_IPSEC_OFFLOAD_RX_SYNDROME_DECRYPTED,
+	MLX5E_IPSEC_OFFLOAD_RX_SYNDROME_AUTH_FAILED,
+	MLX5E_IPSEC_OFFLOAD_RX_SYNDROME_BAD_TRAILER,
+};
+
+void mlx5e_ipsec_offload_handle_rx_skb(struct net_device *netdev,
+				       struct sk_buff *skb,
+				       struct mlx5_cqe64 *cqe)
 {
-	struct mlx5e_priv *priv = netdev_priv(netdev);
+	u32 ipsec_meta_data = be32_to_cpu(cqe->ft_metadata);
+	struct mlx5e_priv *priv;
 	struct xfrm_offload *xo;
 	struct xfrm_state *xs;
 	struct sec_path *sp;
-	u32 sa_handle;
+	u32  sa_handle;
 
+	sa_handle = MLX5_IPSEC_METADATA_HANDLE(ipsec_meta_data);
+	priv = netdev_priv(netdev);
 	sp = secpath_set(skb);
 	if (unlikely(!sp)) {
 		atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_sp_alloc);
-		return NULL;
+		return;
 	}
 
-	sa_handle = be32_to_cpu(mdata->content.rx.sa_handle);
 	xs = mlx5e_ipsec_sadb_rx_lookup(priv->ipsec, sa_handle);
 	if (unlikely(!xs)) {
 		atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_sadb_miss);
-		return NULL;
+		return;
 	}
 
-	sp = skb_sec_path(skb);
 	sp->xvec[sp->len++] = xs;
 	sp->olen++;
 
 	xo = xfrm_offload(skb);
 	xo->flags = CRYPTO_DONE;
-	switch (mdata->syndrome) {
-	case MLX5E_IPSEC_RX_SYNDROME_DECRYPTED:
+
+	switch (MLX5_IPSEC_METADATA_SYNDROM(ipsec_meta_data)) {
+	case MLX5E_IPSEC_OFFLOAD_RX_SYNDROME_DECRYPTED:
 		xo->status = CRYPTO_SUCCESS;
-		if (likely(priv->ipsec->no_trailer)) {
-			xo->flags |= XFRM_ESP_NO_TRAILER;
-			xo->proto = mdata->content.rx.nexthdr;
-		}
 		break;
-	case MLX5E_IPSEC_RX_SYNDROME_AUTH_FAILED:
+	case MLX5E_IPSEC_OFFLOAD_RX_SYNDROME_AUTH_FAILED:
 		xo->status = CRYPTO_TUNNEL_ESP_AUTH_FAILED;
 		break;
-	case MLX5E_IPSEC_RX_SYNDROME_BAD_PROTO:
-		xo->status = CRYPTO_INVALID_PROTOCOL;
+	case MLX5E_IPSEC_OFFLOAD_RX_SYNDROME_BAD_TRAILER:
+		xo->status = CRYPTO_INVALID_PACKET_SYNTAX;
 		break;
 	default:
 		atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_syndrome);
-		return NULL;
-	}
-	return xs;
-}
-
-struct sk_buff *mlx5e_ipsec_handle_rx_skb(struct net_device *netdev,
-					  struct sk_buff *skb, u32 *cqe_bcnt)
-{
-	struct mlx5e_ipsec_metadata *mdata;
-	struct xfrm_state *xs;
-
-	if (!is_metadata_hdr_valid(skb))
-		return skb;
-
-	/* Use the metadata */
-	mdata = (struct mlx5e_ipsec_metadata *)(skb->data + ETH_HLEN);
-	xs = mlx5e_ipsec_build_sp(netdev, skb, mdata);
-	if (unlikely(!xs)) {
-		kfree_skb(skb);
-		return NULL;
-	}
-
-	remove_metadata_hdr(skb);
-	*cqe_bcnt -= MLX5E_METADATA_ETHER_LEN;
-
-	return skb;
-}
-
-bool mlx5e_ipsec_feature_check(struct sk_buff *skb, struct net_device *netdev,
-			       netdev_features_t features)
-{
-	struct sec_path *sp = skb_sec_path(skb);
-	struct xfrm_state *x;
-
-	if (sp && sp->len) {
-		x = sp->xvec[0];
-		if (x && x->xso.offload_handle)
-			return true;
-	}
-	return false;
-}
-
-void mlx5e_ipsec_build_inverse_table(void)
-{
-	u16 mss_inv;
-	u32 mss;
-
-	/* Calculate 1/x inverse table for use in GSO data path.
-	 * Using this table, we provide the IPSec accelerator with the value of
-	 * 1/gso_size so that it can infer the position of each segment inside
-	 * the GSO, and increment the ESP sequence number, and generate the IV.
-	 * The HW needs this value in Q0.16 fixed-point number format
-	 */
-	mlx5e_ipsec_inverse_table[1] = htons(0xFFFF);
-	for (mss = 2; mss < MAX_LSO_MSS; mss++) {
-		mss_inv = div_u64(1ULL << 32, mss) >> 16;
-		mlx5e_ipsec_inverse_table[mss] = htons(mss_inv);
 	}
 }