* [MPTCP] [PATCH] Revert tcp_skb_cb to it's original size
@ 2017-05-03 22:02 Rao Shoaib
0 siblings, 0 replies; only message in thread
From: Rao Shoaib @ 2017-05-03 22:02 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 23874 bytes --]
This is an attempt to address the issue without allocating any space and overloading existing data structures.
Shoaib.
Signed-off-by: Rao Shoaib <rao.shoaib(a)oracle.com>
---
include/linux/skbuff.h | 8 ++-
include/net/mptcp.h | 17 +++---
include/net/tcp.h | 19 +++---
net/ipv4/tcp_ipv4.c | 4 +-
net/ipv4/tcp_output.c | 13 ++++-
net/ipv6/tcp_ipv6.c | 4 +-
net/mptcp/mptcp_input.c | 37 ++++++------
net/mptcp/mptcp_ipv4.c | 4 +-
net/mptcp/mptcp_ipv6.c | 4 +-
net/mptcp/mptcp_output.c | 146 +++++++++++++++++------------------------------
10 files changed, 113 insertions(+), 143 deletions(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f66cd5e..e2cdc21 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -532,7 +532,13 @@ struct sk_buff {
struct rb_node rbnode; /* used in netem & tcp stack */
};
struct sock *sk;
- struct net_device *dev;
+ union {
+ struct net_device *dev;
+ struct {
+ __u8 mptcp_flags;
+ __u8 dss_offset;
+ };
+ };
/*
* This is the control buffer. It is free to use for every
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index 876f1e6..d9ff2e7 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -807,8 +807,7 @@ void tcp_parse_mptcp_options(const struct sk_buff *skb,
struct mptcp_options_received *mopt);
void mptcp_parse_options(const uint8_t *ptr, int opsize,
struct mptcp_options_received *mopt,
- const struct sk_buff *skb,
- struct tcp_sock *tp);
+ struct sk_buff *skb, struct tcp_sock *tp);
void mptcp_syn_options(const struct sock *sk, struct tcp_out_options *opts,
unsigned *remaining);
void mptcp_synack_options(struct request_sock *req,
@@ -1000,12 +999,12 @@ static inline void mptcp_sub_force_close_all(struct mptcp_cb *mpcb,
static inline bool mptcp_is_data_seq(const struct sk_buff *skb)
{
- return TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_SEQ;
+ return skb->mptcp_flags & MPTCPHDR_SEQ;
}
static inline bool mptcp_is_data_fin(const struct sk_buff *skb)
{
- return TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_FIN;
+ return skb->mptcp_flags & MPTCPHDR_FIN;
}
/* Is it a data-fin while in infinite mapping mode?
@@ -1034,17 +1033,17 @@ static inline u8 mptcp_get_64_bit(u64 data_seq, struct mptcp_cb *mpcb)
/* Sets the data_seq and returns pointer to the in-skb field of the data_seq.
* If the packet has a 64-bit dseq, the pointer points to the last 32 bits.
*/
-static inline __u32 *mptcp_skb_set_data_seq(const struct sk_buff *skb,
+static inline __u32 *mptcp_skb_set_data_seq(struct sk_buff *skb,
u32 *data_seq,
struct mptcp_cb *mpcb)
{
- __u32 *ptr = (__u32 *)(skb_transport_header(skb) + TCP_SKB_CB(skb)->dss_off);
+ __u32 *ptr = (__u32 *)(skb_transport_header(skb) + skb->dss_offset);
- if (TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_SEQ64_SET) {
+ if (skb->mptcp_flags & MPTCPHDR_SEQ64_SET) {
u64 data_seq64 = get_unaligned_be64(ptr);
if (mpcb)
- TCP_SKB_CB(skb)->mptcp_flags |= mptcp_get_64_bit(data_seq64, mpcb);
+ skb->mptcp_flags |= mptcp_get_64_bit(data_seq64, mpcb);
*data_seq = (u32)data_seq64;
ptr++;
@@ -1142,7 +1141,7 @@ static inline void mptcp_reset_mopt(struct tcp_sock *tp)
static inline __be32 mptcp_get_highorder_sndbits(const struct sk_buff *skb,
const struct mptcp_cb *mpcb)
{
- return htonl(mpcb->snd_high_order[(TCP_SKB_CB(skb)->mptcp_flags &
+ return htonl(mpcb->snd_high_order[(skb->mptcp_flags &
MPTCPHDR_SEQ64_INDEX) ? 1 : 0]);
}
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 655ecd4..3258721 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -842,11 +842,6 @@ struct tcp_skb_cb {
__u32 tcp_gso_segs;
};
-#ifdef CONFIG_MPTCP
- __u8 mptcp_flags; /* flags for the MPTCP layer */
- __u8 dss_off; /* Number of 4-byte words until
- * seq-number */
-#endif
__u8 tcp_flags; /* TCP header flags. (tcp[13]) */
__u8 sacked; /* State flags for SACK/FACK. */
@@ -861,7 +856,13 @@ struct tcp_skb_cb {
__u8 ip_dsfield; /* IPv4 tos or IPv6 dsfield */
/* 1 byte hole */
- __u32 ack_seq; /* Sequence number ACK'd */
+ union {
+ __u32 ack_seq; /* Sequence number ACK'd */
+ union {
+ __u32 mptcp_data_seq;
+ __u32 path_mask;
+ };
+ };
union {
union {
struct inet_skb_parm h4;
@@ -869,12 +870,6 @@ struct tcp_skb_cb {
struct inet6_skb_parm h6;
#endif
} header; /* For incoming frames */
-#ifdef CONFIG_MPTCP
- union { /* For MPTCP outgoing frames */
- __u32 path_mask; /* paths that tried to send this skb */
- __u32 dss[6]; /* DSS options */
- };
-#endif
};
};
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 6967a86..4e8a3e3 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1637,8 +1637,8 @@ int tcp_v4_rcv(struct sk_buff *skb)
skb->len - th->doff * 4);
TCP_SKB_CB(skb)->ack_seq = ntohl(th->ack_seq);
#ifdef CONFIG_MPTCP
- TCP_SKB_CB(skb)->mptcp_flags = 0;
- TCP_SKB_CB(skb)->dss_off = 0;
+ skb->mptcp_flags = 0;
+ skb->dss_offset = 0;
#endif
TCP_SKB_CB(skb)->tcp_flags = tcp_flag_byte(th);
TCP_SKB_CB(skb)->tcp_tw_isn = 0;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index a635483..9d49125 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -999,7 +999,6 @@ int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
th->source = inet->inet_sport;
th->dest = inet->inet_dport;
th->seq = htonl(tcb->seq);
- th->ack_seq = htonl(tp->rcv_nxt);
*(((__be16 *)th) + 6) = htons(((tcp_header_size >> 2) << 12) |
tcb->tcp_flags);
@@ -1029,6 +1028,12 @@ int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
if (likely((tcb->tcp_flags & TCPHDR_SYN) == 0))
tcp_ecn_send(sk, skb, tcp_header_size);
+ /*
+ * It is important that we initialize ack_seq after options
+ * have been added as ack_seq is used to pass options value to MPTCP
+ */
+ th->ack_seq = htonl(tp->rcv_nxt);
+
#ifdef CONFIG_TCP_MD5SIG
/* Calculate the MD5 hash, as we have all we need now */
if (md5) {
@@ -1216,6 +1221,12 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len,
TCP_SKB_CB(skb)->tcp_flags = flags & ~(TCPHDR_FIN | TCPHDR_PSH);
TCP_SKB_CB(buff)->tcp_flags = flags;
TCP_SKB_CB(buff)->sacked = TCP_SKB_CB(skb)->sacked;
+ if (mptcp(tp)) {
+ /*
+ * update the mapping
+ */
+ TCP_SKB_CB(buff)->mptcp_data_seq += len;
+ }
if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_PARTIAL) {
/* Copy and checksum data tail into the new buffer. */
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index eba2436..e0c6132 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1413,8 +1413,8 @@ static void tcp_v6_fill_cb(struct sk_buff *skb, const struct ipv6hdr *hdr,
skb->len - th->doff*4);
TCP_SKB_CB(skb)->ack_seq = ntohl(th->ack_seq);
#ifdef CONFIG_MPTCP
- TCP_SKB_CB(skb)->mptcp_flags = 0;
- TCP_SKB_CB(skb)->dss_off = 0;
+ skb->mptcp_flags = 0;
+ skb->dss_offset = 0;
#endif
TCP_SKB_CB(skb)->tcp_flags = tcp_flag_byte(th);
TCP_SKB_CB(skb)->tcp_tw_isn = 0;
diff --git a/net/mptcp/mptcp_input.c b/net/mptcp/mptcp_input.c
index 51cbb06..3c267ad 100644
--- a/net/mptcp/mptcp_input.c
+++ b/net/mptcp/mptcp_input.c
@@ -324,8 +324,8 @@ static int mptcp_verif_dss_csum(struct sock *sk)
* in the final csum_partial-call.
*/
u32 offset = skb_transport_offset(tmp) +
- TCP_SKB_CB(tmp)->dss_off;
- if (TCP_SKB_CB(tmp)->mptcp_flags & MPTCPHDR_SEQ64_SET)
+ tmp->dss_offset;
+ if (tmp->mptcp_flags & MPTCPHDR_SEQ64_SET)
offset += 4;
csum_tcp = skb_checksum(tmp, offset,
@@ -758,7 +758,7 @@ static int mptcp_detect_mapping(struct sock *sk, struct sk_buff *skb)
}
/* Does the DSS had 64-bit seqnum's ? */
- if (!(tcb->mptcp_flags & MPTCPHDR_SEQ64_SET)) {
+ if (!(skb->mptcp_flags & MPTCPHDR_SEQ64_SET)) {
/* Wrapped around? */
if (unlikely(after(data_seq, meta_tp->rcv_nxt) && data_seq < meta_tp->rcv_nxt)) {
tp->mptcp->map_data_seq = mptcp_get_data_seq_64(mpcb, !mpcb->rcv_hiseq_index, data_seq);
@@ -767,9 +767,9 @@ static int mptcp_detect_mapping(struct sock *sk, struct sk_buff *skb)
tp->mptcp->map_data_seq = mptcp_get_data_seq_64(mpcb, mpcb->rcv_hiseq_index, data_seq);
}
} else {
- tp->mptcp->map_data_seq = mptcp_get_data_seq_64(mpcb, (tcb->mptcp_flags & MPTCPHDR_SEQ64_INDEX) ? 1 : 0, data_seq);
+ tp->mptcp->map_data_seq = mptcp_get_data_seq_64(mpcb, (skb->mptcp_flags & MPTCPHDR_SEQ64_INDEX) ? 1 : 0, data_seq);
- if (unlikely(tcb->mptcp_flags & MPTCPHDR_SEQ64_OFO)) {
+ if (unlikely(skb->mptcp_flags & MPTCPHDR_SEQ64_OFO)) {
/* We make sure that the data_seq is invalid.
* It will be dropped later.
*/
@@ -1108,7 +1108,7 @@ int mptcp_check_req(struct sk_buff *skb, struct net *net)
if (!meta_sk)
return 0;
- TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_JOIN;
+ skb->mptcp_flags |= MPTCPHDR_JOIN;
bh_lock_sock_nested(meta_sk);
if (sock_owned_by_user(meta_sk)) {
@@ -1222,7 +1222,7 @@ int mptcp_lookup_join(struct sk_buff *skb, struct inet_timewait_sock *tw)
inet_twsk_put(tw);
}
- TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_JOIN;
+ skb->mptcp_flags |= MPTCPHDR_JOIN;
/* OK, this is a new syn/join, let's create a new open request and
* send syn+ack
*/
@@ -1279,7 +1279,7 @@ int mptcp_do_join_short(struct sk_buff *skb,
return -1;
}
- TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_JOIN;
+ skb->mptcp_flags |= MPTCPHDR_JOIN;
/* OK, this is a new syn/join, let's create a new open request and
* send syn+ack
@@ -1461,7 +1461,7 @@ static void mptcp_data_ack(struct sock *sk, const struct sk_buff *skb)
{
struct sock *meta_sk = mptcp_meta_sk(sk);
struct tcp_sock *meta_tp = tcp_sk(meta_sk), *tp = tcp_sk(sk);
- struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
+ //struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
u32 prior_snd_una = meta_tp->snd_una;
int prior_packets;
u32 nwin, data_ack, data_seq;
@@ -1481,7 +1481,7 @@ static void mptcp_data_ack(struct sock *sk, const struct sk_buff *skb)
/* If we are in infinite mapping mode, rx_opt.data_ack has been
* set by mptcp_clean_rtx_infinite.
*/
- if (!(tcb->mptcp_flags & MPTCPHDR_ACK) && !tp->mpcb->infinite_mapping_snd)
+ if (!(skb->mptcp_flags & MPTCPHDR_ACK) && !tp->mpcb->infinite_mapping_snd)
goto exit;
data_ack = tp->mptcp->rx_opt.data_ack;
@@ -1647,7 +1647,7 @@ static inline bool is_valid_addropt_opsize(u8 mptcp_ver,
void mptcp_parse_options(const uint8_t *ptr, int opsize,
struct mptcp_options_received *mopt,
- const struct sk_buff *skb,
+ struct sk_buff *skb,
struct tcp_sock *tp)
{
const struct mptcp_option *mp_opt = (struct mptcp_option *)ptr;
@@ -1757,7 +1757,7 @@ void mptcp_parse_options(const uint8_t *ptr, int opsize,
ptr += 4;
if (mdss->A) {
- tcb->mptcp_flags |= MPTCPHDR_ACK;
+ skb->mptcp_flags |= MPTCPHDR_ACK;
if (mdss->a) {
mopt->data_ack = (u32) get_unaligned_be64(ptr);
@@ -1768,13 +1768,14 @@ void mptcp_parse_options(const uint8_t *ptr, int opsize,
}
}
- tcb->dss_off = (ptr - skb_transport_header(skb));
+ //tcb->dss_off = (ptr - skb_transport_header(skb));
+ skb->dss_offset = (ptr - skb_transport_header(skb));
if (mdss->M) {
if (mdss->m) {
u64 data_seq64 = get_unaligned_be64(ptr);
- tcb->mptcp_flags |= MPTCPHDR_SEQ64_SET;
+ skb->mptcp_flags |= MPTCPHDR_SEQ64_SET;
mopt->data_seq = (u32) data_seq64;
ptr += 12; /* 64-bit dseq + subseq */
@@ -1784,15 +1785,15 @@ void mptcp_parse_options(const uint8_t *ptr, int opsize,
}
mopt->data_len = get_unaligned_be16(ptr);
- tcb->mptcp_flags |= MPTCPHDR_SEQ;
+ skb->mptcp_flags |= MPTCPHDR_SEQ;
/* Is a check-sum present? */
if (opsize == mptcp_sub_len_dss(mdss, 1))
- tcb->mptcp_flags |= MPTCPHDR_DSS_CSUM;
+ skb->mptcp_flags |= MPTCPHDR_DSS_CSUM;
/* DATA_FIN only possible with DSS-mapping */
if (mdss->F)
- tcb->mptcp_flags |= MPTCPHDR_FIN;
+ skb->mptcp_flags |= MPTCPHDR_FIN;
}
break;
@@ -2185,7 +2186,7 @@ bool mptcp_handle_options(struct sock *sk, const struct tcphdr *th,
* receiver MUST close the subflow with a RST as it is considered broken.
*/
if (mptcp_is_data_seq(skb) && tp->mpcb->dss_csum &&
- !(TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_DSS_CSUM)) {
+ !(skb->mptcp_flags & MPTCPHDR_DSS_CSUM)) {
mptcp_send_reset(sk);
return true;
}
diff --git a/net/mptcp/mptcp_ipv4.c b/net/mptcp/mptcp_ipv4.c
index a147b20..1614836 100644
--- a/net/mptcp/mptcp_ipv4.c
+++ b/net/mptcp/mptcp_ipv4.c
@@ -186,7 +186,7 @@ int mptcp_v4_do_rcv(struct sock *meta_sk, struct sk_buff *skb)
struct sock *child, *rsk = NULL;
int ret;
- if (!(TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_JOIN)) {
+ if (!(skb->mptcp_flags & MPTCPHDR_JOIN)) {
struct tcphdr *th = tcp_hdr(skb);
const struct iphdr *iph = ip_hdr(skb);
struct sock *sk;
@@ -217,7 +217,7 @@ int mptcp_v4_do_rcv(struct sock *meta_sk, struct sk_buff *skb)
return ret;
}
- TCP_SKB_CB(skb)->mptcp_flags = 0;
+ skb->mptcp_flags = 0;
/* Has been removed from the tk-table. Thus, no new subflows.
*
diff --git a/net/mptcp/mptcp_ipv6.c b/net/mptcp/mptcp_ipv6.c
index 0de953d..1959d01 100644
--- a/net/mptcp/mptcp_ipv6.c
+++ b/net/mptcp/mptcp_ipv6.c
@@ -199,7 +199,7 @@ int mptcp_v6_do_rcv(struct sock *meta_sk, struct sk_buff *skb)
struct sock *child, *rsk = NULL;
int ret;
- if (!(TCP_SKB_CB(skb)->mptcp_flags & MPTCPHDR_JOIN)) {
+ if (!(skb->mptcp_flags & MPTCPHDR_JOIN)) {
struct tcphdr *th = tcp_hdr(skb);
const struct ipv6hdr *ip6h = ipv6_hdr(skb);
struct sock *sk;
@@ -232,7 +232,7 @@ int mptcp_v6_do_rcv(struct sock *meta_sk, struct sk_buff *skb)
return ret;
}
- TCP_SKB_CB(skb)->mptcp_flags = 0;
+ skb->mptcp_flags = 0;
/* Has been removed from the tk-table. Thus, no new subflows.
*
diff --git a/net/mptcp/mptcp_output.c b/net/mptcp/mptcp_output.c
index 691ef6f..e746565 100644
--- a/net/mptcp/mptcp_output.c
+++ b/net/mptcp/mptcp_output.c
@@ -59,39 +59,17 @@ EXPORT_SYMBOL(mptcp_sub_len_remove_addr_align);
*/
static bool mptcp_reconstruct_mapping(struct sk_buff *skb)
{
- const struct mp_dss *mpdss = (struct mp_dss *)TCP_SKB_CB(skb)->dss;
- u32 *p32;
- u16 *p16;
-
if (!mptcp_is_data_seq(skb))
return false;
- if (!mpdss->M)
- return false;
-
- /* Move the pointer to the data-seq */
- p32 = (u32 *)mpdss;
- p32++;
- if (mpdss->A) {
- p32++;
- if (mpdss->a)
- p32++;
- }
-
- TCP_SKB_CB(skb)->seq = ntohl(*p32);
-
- /* Get the data_len to calculate the end_data_seq */
- p32++;
- p32++;
- p16 = (u16 *)p32;
- TCP_SKB_CB(skb)->end_seq = ntohs(*p16) + TCP_SKB_CB(skb)->seq;
+ TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->mptcp_data_seq;
return true;
}
static bool mptcp_is_reinjected(const struct sk_buff *skb)
{
- return TCP_SKB_CB(skb)->mptcp_flags & MPTCP_REINJECT;
+ return skb->mptcp_flags & MPTCP_REINJECT;
}
static void mptcp_find_and_set_pathmask(const struct sock *meta_sk, struct sk_buff *skb)
@@ -182,7 +160,7 @@ static void __mptcp_reinject_data(struct sk_buff *orig_skb, struct sock *meta_sk
/* Segment goes back to the MPTCP-layer. So, we need to zero the
* path_mask/dss.
*/
- memset(TCP_SKB_CB(skb)->dss, 0 , mptcp_dss_len);
+ TCP_SKB_CB(skb)->path_mask = 0;
/* We need to find out the path-mask from the meta-write-queue
* to properly select a subflow.
@@ -272,7 +250,7 @@ void mptcp_reinject_data(struct sock *sk, int clone_it)
if (mptcp_is_reinjected(skb_it))
continue;
- tcb->mptcp_flags |= MPTCP_REINJECT;
+ skb_it->mptcp_flags |= MPTCP_REINJECT;
__mptcp_reinject_data(skb_it, meta_sk, sk, clone_it);
}
@@ -319,25 +297,45 @@ combine:
}
}
-static int mptcp_write_dss_mapping(const struct tcp_sock *tp, const struct sk_buff *skb,
- __be32 *ptr)
+/*
+ * RFC6824 states that once a particular subflow mapping has been sent
+ * out it must never be changed. However, packets may be split while
+ * they are in the retransmission queue (due to SACK or ACKs) and that
+ * arguably means that we would change the mapping (e.g. it splits it,
+ * our sends out a subset of the initial mapping).
+ *
+ * Furthermore, the skb checksum is not always preserved across splits
+ * (e.g. mptcp_fragment) which would mean that we need to recompute
+ * the DSS checksum in this case.
+ *
+ * To avoid this we save the initial DSS mapping which allows us to
+ * send the same DSS mapping even for fragmented retransmits.
+ */
+
+static int mptcp_write_dss_mapping(const struct tcp_sock *tp,
+ const struct sk_buff *skb, __be32 *ptr)
{
const struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
- __be32 *start = ptr;
__u16 data_len;
- *ptr++ = htonl(tcb->seq); /* data_seq */
+ *ptr++ = htonl(tcb->mptcp_data_seq); /* data_seq */
/* If it's a non-data DATA_FIN, we set subseq to 0 (draft v7) */
if (mptcp_is_data_fin(skb) && skb->len == 0)
*ptr++ = 0; /* subseq */
else
- *ptr++ = htonl(tp->write_seq - tp->mptcp->snt_isn); /* subseq */
+ *ptr++ = htonl(tcb->seq - tp->mptcp->snt_isn); /* subseq */
- if (tcb->mptcp_flags & MPTCPHDR_INF)
+ if (skb->mptcp_flags & MPTCPHDR_INF)
data_len = 0;
- else
+ else {
data_len = tcb->end_seq - tcb->seq;
+ /*
+ * mptcp_entail_skb adds one for FIN
+ */
+ if (tcb->tcp_flags & TCPHDR_FIN)
+ data_len -= 1;
+ }
if (tp->mpcb->dss_csum && data_len) {
__be16 *p16 = (__be16 *)ptr;
@@ -356,11 +354,11 @@ static int mptcp_write_dss_mapping(const struct tcp_sock *tp, const struct sk_bu
(TCPOPT_NOP));
}
- return ptr - start;
+ return mptcp_dss_len/sizeof(*ptr);
}
-static int mptcp_write_dss_data_ack(const struct tcp_sock *tp, const struct sk_buff *skb,
- __be32 *ptr)
+static int mptcp_write_dss_data_ack(const struct tcp_sock *tp,
+ const struct sk_buff *skb, __be32 *ptr)
{
struct mp_dss *mdss = (struct mp_dss *)ptr;
__be32 *start = ptr;
@@ -377,54 +375,12 @@ static int mptcp_write_dss_data_ack(const struct tcp_sock *tp, const struct sk_b
mdss->len = mptcp_sub_len_dss(mdss, tp->mpcb->dss_csum);
ptr++;
+ /* data_ack */
*ptr++ = htonl(mptcp_meta_tp(tp)->rcv_nxt);
return ptr - start;
}
-/* RFC6824 states that once a particular subflow mapping has been sent
- * out it must never be changed. However, packets may be split while
- * they are in the retransmission queue (due to SACK or ACKs) and that
- * arguably means that we would change the mapping (e.g. it splits it,
- * our sends out a subset of the initial mapping).
- *
- * Furthermore, the skb checksum is not always preserved across splits
- * (e.g. mptcp_fragment) which would mean that we need to recompute
- * the DSS checksum in this case.
- *
- * To avoid this we save the initial DSS mapping which allows us to
- * send the same DSS mapping even for fragmented retransmits.
- */
-static void mptcp_save_dss_data_seq(const struct tcp_sock *tp, struct sk_buff *skb)
-{
- struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
- __be32 *ptr = (__be32 *)tcb->dss;
-
- tcb->mptcp_flags |= MPTCPHDR_SEQ;
-
- ptr += mptcp_write_dss_data_ack(tp, skb, ptr);
- ptr += mptcp_write_dss_mapping(tp, skb, ptr);
-}
-
-/* Write the saved DSS mapping to the header */
-static int mptcp_write_dss_data_seq(const struct tcp_sock *tp, struct sk_buff *skb,
- __be32 *ptr)
-{
- __be32 *start = ptr;
-
- memcpy(ptr, TCP_SKB_CB(skb)->dss, mptcp_dss_len);
-
- /* update the data_ack */
- start[1] = htonl(mptcp_meta_tp(tp)->rcv_nxt);
-
- /* dss is in a union with inet_skb_parm and
- * the IP layer expects zeroed IPCB fields.
- */
- memset(TCP_SKB_CB(skb)->dss, 0 , mptcp_dss_len);
-
- return mptcp_dss_len/sizeof(*ptr);
-}
-
static bool mptcp_skb_entail(struct sock *sk, struct sk_buff *skb, int reinject)
{
struct tcp_sock *tp = tcp_sk(sk);
@@ -434,7 +390,7 @@ static bool mptcp_skb_entail(struct sock *sk, struct sk_buff *skb, int reinject)
struct sk_buff *subskb = NULL;
if (!reinject)
- TCP_SKB_CB(skb)->mptcp_flags |= (mpcb->snd_hiseq_index ?
+ skb->mptcp_flags |= (mpcb->snd_hiseq_index ?
MPTCPHDR_SEQ64_INDEX : 0);
subskb = pskb_copy_for_clone(skb, GFP_ATOMIC);
@@ -463,14 +419,15 @@ static bool mptcp_skb_entail(struct sock *sk, struct sk_buff *skb, int reinject)
tp->mptcp->fully_established = 1;
tp->mpcb->infinite_mapping_snd = 1;
tp->mptcp->infinite_cutoff_seq = tp->write_seq;
- tcb->mptcp_flags |= MPTCPHDR_INF;
+ skb->mptcp_flags |= MPTCPHDR_INF;
}
if (mptcp_is_data_fin(subskb))
mptcp_combine_dfin(subskb, meta_sk, sk);
- mptcp_save_dss_data_seq(tp, subskb);
-
+ skb->mptcp_flags |= MPTCPHDR_SEQ;
+ tcb->mptcp_data_seq = tcb->seq;
+
tcb->seq = tp->write_seq;
/* Take into account seg len */
@@ -536,9 +493,9 @@ static int mptcp_fragment(struct sock *meta_sk, struct sk_buff *skb, u32 len,
buff = skb->next;
- flags = TCP_SKB_CB(skb)->mptcp_flags;
- TCP_SKB_CB(skb)->mptcp_flags = flags & ~(MPTCPHDR_FIN);
- TCP_SKB_CB(buff)->mptcp_flags = flags;
+ flags = skb->mptcp_flags;
+ skb->mptcp_flags = flags & ~(MPTCPHDR_FIN);
+ buff->mptcp_flags = flags;
TCP_SKB_CB(buff)->path_mask = TCP_SKB_CB(skb)->path_mask;
/* If reinject == 1, the buff will be added to the reinject
@@ -554,7 +511,8 @@ static int mptcp_fragment(struct sock *meta_sk, struct sk_buff *skb, u32 len,
tcp_sk(meta_sk)->mpcb->reinject_queue.qlen++;
meta_sk->sk_write_queue.qlen--;
- if (!before(tcp_sk(meta_sk)->snd_nxt, TCP_SKB_CB(buff)->end_seq)) {
+ if (!before(tcp_sk(meta_sk)->snd_nxt,
+ TCP_SKB_CB(buff)->end_seq)) {
undo = old_factor - tcp_skb_pcount(skb) -
tcp_skb_pcount(buff);
if (undo)
@@ -967,7 +925,7 @@ void mptcp_established_options(struct sock *sk, struct sk_buff *skb,
if (unlikely(mpcb->infinite_mapping_snd) &&
((mpcb->send_infinite_mapping && tcb &&
mptcp_is_data_seq(skb) &&
- !(tcb->mptcp_flags & MPTCPHDR_INF) &&
+ !(skb->mptcp_flags & MPTCPHDR_INF) &&
!before(tcb->seq, tp->mptcp->infinite_cutoff_seq)) ||
!mpcb->send_infinite_mapping))
return;
@@ -1197,10 +1155,10 @@ void mptcp_options_write(__be32 *ptr, struct tcp_sock *tp,
}
if (OPTION_DATA_ACK & opts->mptcp_options) {
- if (!mptcp_is_data_seq(skb))
- ptr += mptcp_write_dss_data_ack(tp, skb, ptr);
- else
- ptr += mptcp_write_dss_data_seq(tp, skb, ptr);
+ ptr += mptcp_write_dss_data_ack(tp, skb, ptr);
+ if (mptcp_is_data_seq(skb))
+ ptr += mptcp_write_dss_mapping(tp, skb, ptr);
+ skb->dev = NULL;
}
if (unlikely(OPTION_MP_PRIO & opts->mptcp_options)) {
struct mp_prio *mpprio = (struct mp_prio *)ptr;
@@ -1233,7 +1191,7 @@ void mptcp_send_fin(struct sock *meta_sk)
mss_now = mptcp_current_mss(meta_sk);
if (tcp_send_head(meta_sk) != NULL) {
- TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_FIN;
+ skb->mptcp_flags |= MPTCPHDR_FIN;
TCP_SKB_CB(skb)->end_seq++;
meta_tp->write_seq++;
} else {
@@ -1250,7 +1208,7 @@ void mptcp_send_fin(struct sock *meta_sk)
tcp_init_nondata_skb(skb, meta_tp->write_seq, TCPHDR_ACK);
TCP_SKB_CB(skb)->end_seq++;
- TCP_SKB_CB(skb)->mptcp_flags |= MPTCPHDR_FIN;
+ skb->mptcp_flags |= MPTCPHDR_FIN;
tcp_queue_skb(meta_sk, skb);
}
__tcp_push_pending_frames(meta_sk, mss_now, TCP_NAGLE_OFF);
--
2.7.4
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2017-05-03 22:02 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-05-03 22:02 [MPTCP] [PATCH] Revert tcp_skb_cb to it's original size Rao Shoaib
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.