From: Jon Maloy <jmaloy@redhat.com>
To: passt-dev@passt.top, sbrivio@redhat.com, lvivier@redhat.com,
dgibson@redhat.com, jmaloy@redhat.com
Subject: [PATCH 3/4] tcp: unify l2 TCPv4 and TCPv6 queues and structures
Date: Fri, 6 Sep 2024 17:34:26 -0400 [thread overview]
Message-ID: <20240906213427.1915806-4-jmaloy@redhat.com> (raw)
In-Reply-To: <20240906213427.1915806-1-jmaloy@redhat.com>
Following the preparations in the previous commits, we can now
remove the queues dedicated for TCPv6 and move that traffic over
to the queues currently used for TCPv4.
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
tcp.c | 8 ++-
tcp_buf.c | 158 +++++++++---------------------------------------------
tcp_buf.h | 1 +
3 files changed, 28 insertions(+), 139 deletions(-)
diff --git a/tcp.c b/tcp.c
index 006e503..19cf9e5 100644
--- a/tcp.c
+++ b/tcp.c
@@ -998,12 +998,14 @@ size_t tcp_l2_buf_fill_headers(const struct tcp_tap_conn *conn,
if (a4) {
iov[TCP_IOV_IP].iov_len = sizeof(struct iphdr);
+ tcp4_eth_src.h_proto = htons_constant(ETH_P_IP);
return tcp_fill_headers4(conn, iov[TCP_IOV_TAP].iov_base,
iov[TCP_IOV_IP].iov_base,
iov[TCP_IOV_PAYLOAD].iov_base, dlen,
check, seq);
} else {
iov[TCP_IOV_IP].iov_len = sizeof(struct ipv6hdr);
+ tcp4_eth_src.h_proto = htons_constant(ETH_P_IPV6);
return tcp_fill_headers6(conn, iov[TCP_IOV_TAP].iov_base,
iov[TCP_IOV_IP].iov_base,
iov[TCP_IOV_PAYLOAD].iov_base, dlen,
@@ -2508,11 +2510,7 @@ int tcp_init(struct ctx *c)
{
ASSERT(!c->no_tcp);
- if (c->ifi4)
- tcp_sock4_iov_init(c);
-
- if (c->ifi6)
- tcp_sock6_iov_init(c);
+ tcp_sock4_iov_init(c);
memset(init_sock_pool4, 0xff, sizeof(init_sock_pool4));
memset(init_sock_pool6, 0xff, sizeof(init_sock_pool6));
diff --git a/tcp_buf.c b/tcp_buf.c
index 6e6549f..92c4d73 100644
--- a/tcp_buf.c
+++ b/tcp_buf.c
@@ -80,7 +80,7 @@ struct tcp_flags_t {
#endif
/* Ethernet header for IPv4 frames */
-static struct ethhdr tcp4_eth_src;
+struct ethhdr tcp4_eth_src;
static struct tap_hdr tcp4_payload_tap_hdr[TCP_FRAMES_MEM];
/* IPv4 headers */
@@ -104,36 +104,14 @@ static struct tcp_flags_t tcp4_flags[TCP_FRAMES_MEM];
static unsigned int tcp4_flags_used;
/* Ethernet header for IPv6 frames */
-static struct ethhdr tcp6_eth_src;
-
-static struct tap_hdr tcp6_payload_tap_hdr[TCP_FRAMES_MEM];
-/* IPv6 headers */
-struct ipv6hdr tcp_payload_ip6;
-static struct iphdr_t tcp6_payload_ip[TCP_FRAMES_MEM];
-/* TCP headers and data for IPv6 frames */
-static struct tcp_payload_t tcp6_payload[TCP_FRAMES_MEM];
-
-static_assert(MSS6 <= sizeof(tcp6_payload[0].data), "MSS6 is greater than 65516");
-
-/* References tracking the owner connection of frames in the tap outqueue */
-static struct tcp_tap_conn *tcp6_frame_conns[TCP_FRAMES_MEM];
-static unsigned int tcp6_payload_used;
-
-static struct tap_hdr tcp6_flags_tap_hdr[TCP_FRAMES_MEM];
-/* IPv6 headers for TCP segment without payload */
-static struct iphdr_t tcp6_flags_ip[TCP_FRAMES_MEM];
-/* TCP segment without payload for IPv6 frames */
-static struct tcp_flags_t tcp6_flags[TCP_FRAMES_MEM];
-
-static unsigned int tcp6_flags_used;
+struct ipv6hdr tcp_payload_ip6;
/* recvmsg()/sendmsg() data for tap */
static struct iovec iov_sock [TCP_FRAMES_MEM + 1];
static struct iovec tcp4_l2_iov [TCP_FRAMES_MEM][TCP_NUM_IOVS];
-static struct iovec tcp6_l2_iov [TCP_FRAMES_MEM][TCP_NUM_IOVS];
static struct iovec tcp4_l2_flags_iov [TCP_FRAMES_MEM][TCP_NUM_IOVS];
-static struct iovec tcp6_l2_flags_iov [TCP_FRAMES_MEM][TCP_NUM_IOVS];
+
/**
* tcp_update_l2_buf() - Update Ethernet header buffers with addresses
* @eth_d: Ethernet destination address, NULL if unchanged
@@ -142,7 +120,6 @@ static struct iovec tcp6_l2_flags_iov [TCP_FRAMES_MEM][TCP_NUM_IOVS];
void tcp_update_l2_buf(const unsigned char *eth_d, const unsigned char *eth_s)
{
eth_update_mac(&tcp4_eth_src, eth_d, eth_s);
- eth_update_mac(&tcp6_eth_src, eth_d, eth_s);
}
/**
@@ -191,61 +168,12 @@ void tcp_sock4_iov_init(const struct ctx *c)
}
}
-/**
- * tcp_sock6_iov_init() - Initialise scatter-gather L2 buffers for IPv6 sockets
- * @c: Execution context
- */
-void tcp_sock6_iov_init(const struct ctx *c)
-{
- struct ipv6hdr ip6 = L2_BUF_IP6_INIT(IPPROTO_TCP);
- struct iovec *iov;
- int i;
-
- tcp6_eth_src.h_proto = htons_constant(ETH_P_IPV6);
- tcp_payload_ip6 = ip6;
-
- for (i = 0; i < ARRAY_SIZE(tcp6_payload); i++) {
- tcp6_payload[i].th.doff = sizeof(struct tcphdr) / 4;
- tcp6_payload[i].th.ack = 1;
- }
-
- for (i = 0; i < ARRAY_SIZE(tcp6_flags); i++) {
- tcp6_flags_ip[i].ip6 = ip6;
- tcp6_flags[i].th.doff = sizeof(struct tcphdr) / 4;
- tcp6_flags[i].th .ack = 1;
- }
-
- for (i = 0; i < TCP_FRAMES_MEM; i++) {
- iov = tcp6_l2_iov[i];
-
- iov[TCP_IOV_TAP] = tap_hdr_iov(c, &tcp6_payload_tap_hdr[i]);
- iov[TCP_IOV_ETH] = IOV_OF_LVALUE(tcp6_eth_src);
- iov[TCP_IOV_IP].iov_base = &tcp6_payload_ip[i];
- iov[TCP_IOV_IP].iov_len = sizeof(tcp6_payload_ip[i].ip6);
- iov[TCP_IOV_PAYLOAD].iov_base = &tcp6_payload[i];
- }
-
- for (i = 0; i < TCP_FRAMES_MEM; i++) {
- iov = tcp6_l2_flags_iov[i];
-
- iov[TCP_IOV_TAP] = tap_hdr_iov(c, &tcp6_flags_tap_hdr[i]);
- iov[TCP_IOV_ETH] = IOV_OF_LVALUE(tcp6_eth_src);
- iov[TCP_IOV_IP].iov_base = &tcp6_flags_ip[i];
- iov[TCP_IOV_IP].iov_len = sizeof(tcp6_flags_ip[i].ip6);
- iov[TCP_IOV_PAYLOAD].iov_base = &tcp6_flags[i];
- }
-}
-
/**
* tcp_flags_flush() - Send out buffers for segments with no data (flags)
* @c: Execution context
*/
void tcp_flags_flush(const struct ctx *c)
{
- tap_send_frames(c, &tcp6_l2_flags_iov[0][0], TCP_NUM_IOVS,
- tcp6_flags_used);
- tcp6_flags_used = 0;
-
tap_send_frames(c, &tcp4_l2_flags_iov[0][0], TCP_NUM_IOVS,
tcp4_flags_used);
tcp4_flags_used = 0;
@@ -287,14 +215,6 @@ void tcp_payload_flush(struct ctx *c)
{
size_t m;
- m = tap_send_frames(c, &tcp6_l2_iov[0][0], TCP_NUM_IOVS,
- tcp6_payload_used);
- if (m != tcp6_payload_used) {
- tcp_revert_seq(c, &tcp6_frame_conns[m], &tcp6_l2_iov[m],
- tcp6_payload_used - m);
- }
- tcp6_payload_used = 0;
-
m = tap_send_frames(c, &tcp4_l2_iov[0][0], TCP_NUM_IOVS,
tcp4_payload_used);
if (m != tcp4_payload_used) {
@@ -321,21 +241,13 @@ int tcp_buf_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
uint32_t seq;
int ret;
- if (CONN_V4(conn))
- iov = tcp4_l2_flags_iov[tcp4_flags_used++];
- else
- iov = tcp6_l2_flags_iov[tcp6_flags_used++];
-
+ iov = tcp4_l2_flags_iov[tcp4_flags_used++];
payload = iov[TCP_IOV_PAYLOAD].iov_base;
-
seq = conn->seq_to_tap;
ret = tcp_prepare_flags(c, conn, flags, &payload->th,
payload->opts, &optlen);
if (ret <= 0) {
- if (CONN_V4(conn))
- tcp4_flags_used--;
- else
- tcp6_flags_used--;
+ tcp4_flags_used--;
return ret;
}
@@ -346,10 +258,7 @@ int tcp_buf_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
struct iovec *dup_iov;
int i;
- if (CONN_V4(conn))
- dup_iov = tcp4_l2_flags_iov[tcp4_flags_used++];
- else
- dup_iov = tcp6_l2_flags_iov[tcp6_flags_used++];
+ dup_iov = tcp4_l2_flags_iov[tcp4_flags_used++];
for (i = 0; i < TCP_NUM_IOVS; i++)
memcpy(dup_iov[i].iov_base, iov[i].iov_base,
@@ -357,13 +266,8 @@ int tcp_buf_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
dup_iov[TCP_IOV_PAYLOAD].iov_len = iov[TCP_IOV_PAYLOAD].iov_len;
}
- if (CONN_V4(conn)) {
- if (tcp4_flags_used > TCP_FRAMES_MEM - 2)
- tcp_flags_flush(c);
- } else {
- if (tcp6_flags_used > TCP_FRAMES_MEM - 2)
- tcp_flags_flush(c);
- }
+ if (tcp4_flags_used > TCP_FRAMES_MEM - 2)
+ tcp_flags_flush(c);
return 0;
}
@@ -379,36 +283,26 @@ int tcp_buf_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags)
static void tcp_data_to_tap(struct ctx *c, struct tcp_tap_conn *conn,
ssize_t dlen, int no_csum, uint32_t seq)
{
+ struct iovec *iov_prev = tcp4_l2_iov[tcp4_payload_used - 1];
+ const uint16_t *check = NULL;
struct iovec *iov;
size_t l4len;
conn->seq_to_tap = seq + dlen;
- if (CONN_V4(conn)) {
- struct iovec *iov_prev = tcp4_l2_iov[tcp4_payload_used - 1];
- const uint16_t *check = NULL;
-
- if (no_csum) {
- struct iphdr *iph = iov_prev[TCP_IOV_IP].iov_base;
- check = &iph->check;
- }
+ if (CONN_V4(conn) && no_csum) {
+ struct iphdr *iph = iov_prev[TCP_IOV_IP].iov_base;
- tcp4_frame_conns[tcp4_payload_used] = conn;
-
- iov = tcp4_l2_iov[tcp4_payload_used++];
- l4len = tcp_l2_buf_fill_headers(conn, iov, dlen, check, seq);
- iov[TCP_IOV_PAYLOAD].iov_len = l4len;
- if (tcp4_payload_used > TCP_FRAMES_MEM - 1)
- tcp_payload_flush(c);
- } else if (CONN_V6(conn)) {
- tcp6_frame_conns[tcp6_payload_used] = conn;
-
- iov = tcp6_l2_iov[tcp6_payload_used++];
- l4len = tcp_l2_buf_fill_headers(conn, iov, dlen, NULL, seq);
- iov[TCP_IOV_PAYLOAD].iov_len = l4len;
- if (tcp6_payload_used > TCP_FRAMES_MEM - 1)
- tcp_payload_flush(c);
+ check = &iph->check;
}
+
+ tcp4_frame_conns[tcp4_payload_used] = conn;
+
+ iov = tcp4_l2_iov[tcp4_payload_used++];
+ l4len = tcp_l2_buf_fill_headers(conn, iov, dlen, check, seq);
+ iov[TCP_IOV_PAYLOAD].iov_len = l4len;
+ if (tcp4_payload_used > TCP_FRAMES_MEM - 1)
+ tcp_payload_flush(c);
}
/**
@@ -472,19 +366,15 @@ int tcp_buf_data_from_sock(struct ctx *c, struct tcp_tap_conn *conn)
mh_sock.msg_iovlen = fill_bufs;
}
- if (( v4 && tcp4_payload_used + fill_bufs > TCP_FRAMES_MEM) ||
- (!v4 && tcp6_payload_used + fill_bufs > TCP_FRAMES_MEM)) {
+ if ((v4 && tcp4_payload_used + fill_bufs > TCP_FRAMES_MEM)) {
tcp_payload_flush(c);
/* Silence Coverity CWE-125 false positive */
- tcp4_payload_used = tcp6_payload_used = 0;
+ tcp4_payload_used = 0;
}
for (i = 0, iov = iov_sock + 1; i < fill_bufs; i++, iov++) {
- if (v4)
- iov->iov_base = &tcp4_payload[tcp4_payload_used + i].data;
- else
- iov->iov_base = &tcp6_payload[tcp6_payload_used + i].data;
+ iov->iov_base = &tcp4_payload[tcp4_payload_used + i].data;
iov->iov_len = mss;
}
if (iov_rem)
diff --git a/tcp_buf.h b/tcp_buf.h
index d3d0d7f..d7cdbaf 100644
--- a/tcp_buf.h
+++ b/tcp_buf.h
@@ -13,6 +13,7 @@ void tcp_payload_flush(struct ctx *c);
int tcp_buf_data_from_sock(struct ctx *c, struct tcp_tap_conn *conn);
int tcp_buf_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags);
+extern struct ethhdr tcp4_eth_src;
extern struct iphdr tcp_payload_ip4;
extern struct ipv6hdr tcp_payload_ip6;
#endif /*TCP_BUF_H */
--
@@ -13,6 +13,7 @@ void tcp_payload_flush(struct ctx *c);
int tcp_buf_data_from_sock(struct ctx *c, struct tcp_tap_conn *conn);
int tcp_buf_send_flag(struct ctx *c, struct tcp_tap_conn *conn, int flags);
+extern struct ethhdr tcp4_eth_src;
extern struct iphdr tcp_payload_ip4;
extern struct ipv6hdr tcp_payload_ip6;
#endif /*TCP_BUF_H */
--
2.45.2
next prev parent reply other threads:[~2024-09-06 21:34 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-06 21:34 [PATCH 0/4] tcp: unify IPv4 and IPv6 l2 tap queues Jon Maloy
2024-09-06 21:34 ` [PATCH 1/4] tcp: create unified struct for IPv4 and IPv6 header Jon Maloy
2024-09-09 1:04 ` David Gibson
2024-09-06 21:34 ` [PATCH 2/4] tcp: update ip address in l2 tap queues on the fly Jon Maloy
2024-09-09 1:08 ` David Gibson
2024-09-06 21:34 ` Jon Maloy [this message]
2024-09-09 1:17 ` [PATCH 3/4] tcp: unify l2 TCPv4 and TCPv6 queues and structures David Gibson
2024-09-06 21:34 ` [PATCH 4/4] tcp: change prefix tcp4_ to tcp_ where applicable Jon Maloy
2024-09-09 8:31 ` [PATCH 0/4] tcp: unify IPv4 and IPv6 l2 tap queues Stefano Brivio
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240906213427.1915806-4-jmaloy@redhat.com \
--to=jmaloy@redhat.com \
--cc=dgibson@redhat.com \
--cc=lvivier@redhat.com \
--cc=passt-dev@passt.top \
--cc=sbrivio@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).