@@ -41,6 +41,47 @@
#define VIRTIO_NET_RX_QUEUE_MIN_SIZE VIRTIO_NET_RX_QUEUE_DEFAULT_SIZE
#define VIRTIO_NET_TX_QUEUE_MIN_SIZE VIRTIO_NET_TX_QUEUE_DEFAULT_SIZE
+#define VIRTIO_NET_IP4_ADDR_SIZE 8 /* ipv4 saddr + daddr */
+
+#define VIRTIO_NET_TCP_FLAG 0x3F
+#define VIRTIO_NET_TCP_HDR_LENGTH 0xF000
+
+/* IPv4 max payload, 16 bits in the header */
+#define VIRTIO_NET_MAX_IP4_PAYLOAD (65535 - sizeof(struct ip_header))
+#define VIRTIO_NET_MAX_TCP_PAYLOAD 65535
+
+/* header length value in ip header without option */
+#define VIRTIO_NET_IP4_HEADER_LENGTH 5
+
+#define VIRTIO_NET_IP6_ADDR_SIZE 32 /* ipv6 saddr + daddr */
+#define VIRTIO_NET_MAX_IP6_PAYLOAD VIRTIO_NET_MAX_TCP_PAYLOAD
+
+/* Purge coalesced packets timer interval, This value affects the performance
+ a lot, and should be tuned carefully, '300000'(300us) is the recommended
+ value to pass the WHQL test, '50000' can gain 2x netperf throughput with
+ tso/gso/gro 'off'. */
+#define VIRTIO_NET_RSC_DEFAULT_INTERVAL 300000
+
+/* temporary until standard header include it */
+#if !defined(VIRTIO_NET_HDR_F_RSC_INFO)
+
+#define VIRTIO_NET_HDR_F_RSC_INFO 4 /* rsc_ext data in csum_ fields */
+#define VIRTIO_NET_F_RSC_EXT 38
+
+static inline __virtio16 *virtio_net_rsc_ext_num_packets(
+ struct virtio_net_hdr *hdr)
+{
+ return &hdr->csum_start;
+}
+
+static inline __virtio16 *virtio_net_rsc_ext_num_dupacks(
+ struct virtio_net_hdr *hdr)
+{
+ return &hdr->csum_offset;
+}
+
+#endif
+
/*
* Calculate the number of bytes up to and including the given 'field' of
* 'container'.
@@ -628,6 +669,7 @@ static uint64_t virtio_net_get_features(VirtIODevice *vdev, uint64_t features,
if (!get_vhost_net(nc->peer)) {
return features;
}
+
features = vhost_net_get_features(get_vhost_net(nc->peer), features);
vdev->backend_features = features;
@@ -701,6 +743,11 @@ static void virtio_net_set_features(VirtIODevice *vdev, uint64_t features)
virtio_has_feature(features,
VIRTIO_F_VERSION_1));
+ n->rsc4_enabled = virtio_has_feature(features, VIRTIO_NET_F_RSC_EXT) &&
+ virtio_has_feature(features, VIRTIO_NET_F_GUEST_TSO4);
+ n->rsc6_enabled = virtio_has_feature(features, VIRTIO_NET_F_RSC_EXT) &&
+ virtio_has_feature(features, VIRTIO_NET_F_GUEST_TSO6);
+
if (n->has_vnet_hdr) {
n->curr_guest_offloads =
virtio_net_guest_offloads_by_features(features);
@@ -781,6 +828,12 @@ static int virtio_net_handle_offloads(VirtIONet *n, uint8_t cmd,
return VIRTIO_NET_ERR;
}
+ n->rsc4_enabled = virtio_has_feature(offloads, VIRTIO_NET_F_RSC_EXT) &&
+ virtio_has_feature(offloads, VIRTIO_NET_F_GUEST_TSO4);
+ n->rsc6_enabled = virtio_has_feature(offloads, VIRTIO_NET_F_RSC_EXT) &&
+ virtio_has_feature(offloads, VIRTIO_NET_F_GUEST_TSO6);
+ virtio_clear_feature(&offloads, VIRTIO_NET_F_RSC_EXT);
+
supported_offloads = virtio_net_supported_guest_offloads(n);
if (offloads & ~supported_offloads) {
return VIRTIO_NET_ERR;
@@ -1292,7 +1345,7 @@ static ssize_t virtio_net_receive_rcu(NetClientState *nc, const uint8_t *buf,
return size;
}
-static ssize_t virtio_net_receive(NetClientState *nc, const uint8_t *buf,
+static ssize_t virtio_net_do_receive(NetClientState *nc, const uint8_t *buf,
size_t size)
{
ssize_t r;
@@ -1303,6 +1356,612 @@ static ssize_t virtio_net_receive(NetClientState *nc, const uint8_t *buf,
return r;
}
+static void virtio_net_rsc_extract_unit4(VirtioNetRscChain *chain,
+ const uint8_t *buf,
+ VirtioNetRscUnit *unit)
+{
+ uint16_t ip_hdrlen;
+ struct ip_header *ip;
+
+ ip = (struct ip_header *)(buf + chain->n->guest_hdr_len
+ + sizeof(struct eth_header));
+ unit->ip = (void *)ip;
+ ip_hdrlen = (ip->ip_ver_len & 0xF) << 2;
+ unit->ip_plen = &ip->ip_len;
+ unit->tcp = (struct tcp_header *)(((uint8_t *)unit->ip) + ip_hdrlen);
+ unit->tcp_hdrlen = (htons(unit->tcp->th_offset_flags) & 0xF000) >> 10;
+ unit->payload = htons(*unit->ip_plen) - ip_hdrlen - unit->tcp_hdrlen;
+}
+
+static void virtio_net_rsc_extract_unit6(VirtioNetRscChain *chain,
+ const uint8_t *buf,
+ VirtioNetRscUnit *unit)
+{
+ struct ip6_header *ip6;
+
+ ip6 = (struct ip6_header *)(buf + chain->n->guest_hdr_len
+ + sizeof(struct eth_header));
+ unit->ip = ip6;
+ unit->ip_plen = &(ip6->ip6_ctlun.ip6_un1.ip6_un1_plen);
+ unit->tcp = (struct tcp_header *)(((uint8_t *)unit->ip)\
+ + sizeof(struct ip6_header));
+ unit->tcp_hdrlen = (htons(unit->tcp->th_offset_flags) & 0xF000) >> 10;
+
+ /* There is a difference between payload lenght in ipv4 and v6,
+ ip header is excluded in ipv6 */
+ unit->payload = htons(*unit->ip_plen) - unit->tcp_hdrlen;
+}
+
+static size_t virtio_net_rsc_drain_seg(VirtioNetRscChain *chain,
+ VirtioNetRscSeg *seg)
+{
+ int ret;
+ struct virtio_net_hdr *h;
+
+ h = (struct virtio_net_hdr *)seg->buf;
+ h->flags = 0;
+ h->gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+ if (seg->is_coalesced) {
+ *virtio_net_rsc_ext_num_packets(h) = seg->packets;
+ *virtio_net_rsc_ext_num_dupacks(h) = seg->dup_ack;
+ h->flags = VIRTIO_NET_HDR_F_RSC_INFO;
+ if (chain->proto == ETH_P_IP) {
+ h->gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+ } else {
+ h->gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+ }
+ }
+
+ ret = virtio_net_do_receive(seg->nc, seg->buf, seg->size);
+ QTAILQ_REMOVE(&chain->buffers, seg, next);
+ g_free(seg->buf);
+ g_free(seg);
+
+ return ret;
+}
+
+static void virtio_net_rsc_purge(void *opq)
+{
+ VirtioNetRscSeg *seg, *rn;
+ VirtioNetRscChain *chain = (VirtioNetRscChain *)opq;
+
+ QTAILQ_FOREACH_SAFE(seg, &chain->buffers, next, rn) {
+ if (virtio_net_rsc_drain_seg(chain, seg) == 0) {
+ chain->stat.purge_failed++;
+ continue;
+ }
+ }
+
+ chain->stat.timer++;
+ if (!QTAILQ_EMPTY(&chain->buffers)) {
+ timer_mod(chain->drain_timer,
+ qemu_clock_get_ns(QEMU_CLOCK_HOST) + chain->n->rsc_timeout);
+ }
+}
+
+static void virtio_net_rsc_cleanup(VirtIONet *n)
+{
+ VirtioNetRscChain *chain, *rn_chain;
+ VirtioNetRscSeg *seg, *rn_seg;
+
+ QTAILQ_FOREACH_SAFE(chain, &n->rsc_chains, next, rn_chain) {
+ QTAILQ_FOREACH_SAFE(seg, &chain->buffers, next, rn_seg) {
+ QTAILQ_REMOVE(&chain->buffers, seg, next);
+ g_free(seg->buf);
+ g_free(seg);
+ }
+
+ timer_del(chain->drain_timer);
+ timer_free(chain->drain_timer);
+ QTAILQ_REMOVE(&n->rsc_chains, chain, next);
+ g_free(chain);
+ }
+}
+
+static void virtio_net_rsc_cache_buf(VirtioNetRscChain *chain,
+ NetClientState *nc,
+ const uint8_t *buf, size_t size)
+{
+ uint16_t hdr_len;
+ VirtioNetRscSeg *seg;
+
+ hdr_len = chain->n->guest_hdr_len;
+ seg = g_malloc(sizeof(VirtioNetRscSeg));
+ seg->buf = g_malloc(hdr_len + sizeof(struct eth_header)
+ + sizeof(struct ip6_header) + VIRTIO_NET_MAX_TCP_PAYLOAD);
+ memcpy(seg->buf, buf, size);
+ seg->size = size;
+ seg->packets = 1;
+ seg->dup_ack = 0;
+ seg->is_coalesced = 0;
+ seg->nc = nc;
+
+ QTAILQ_INSERT_TAIL(&chain->buffers, seg, next);
+ chain->stat.cache++;
+
+ switch (chain->proto) {
+ case ETH_P_IP:
+ virtio_net_rsc_extract_unit4(chain, seg->buf, &seg->unit);
+ break;
+ case ETH_P_IPV6:
+ virtio_net_rsc_extract_unit6(chain, seg->buf, &seg->unit);
+ break;
+ default:
+ g_assert_not_reached();
+ }
+}
+
+static int32_t virtio_net_rsc_handle_ack(VirtioNetRscChain *chain,
+ VirtioNetRscSeg *seg,
+ const uint8_t *buf,
+ struct tcp_header *n_tcp,
+ struct tcp_header *o_tcp)
+{
+ uint32_t nack, oack;
+ uint16_t nwin, owin;
+
+ nack = htonl(n_tcp->th_ack);
+ nwin = htons(n_tcp->th_win);
+ oack = htonl(o_tcp->th_ack);
+ owin = htons(o_tcp->th_win);
+
+ if ((nack - oack) >= VIRTIO_NET_MAX_TCP_PAYLOAD) {
+ chain->stat.ack_out_of_win++;
+ return RSC_FINAL;
+ } else if (nack == oack) {
+ /* duplicated ack or window probe */
+ if (nwin == owin) {
+ /* duplicated ack, add dup ack count due to whql test up to 1 */
+ chain->stat.dup_ack++;
+ return RSC_FINAL;
+ } else {
+ /* Coalesce window update */
+ o_tcp->th_win = n_tcp->th_win;
+ chain->stat.win_update++;
+ return RSC_COALESCE;
+ }
+ } else {
+ /* pure ack, go to 'C', finalize*/
+ chain->stat.pure_ack++;
+ return RSC_FINAL;
+ }
+}
+
+static int32_t virtio_net_rsc_coalesce_data(VirtioNetRscChain *chain,
+ VirtioNetRscSeg *seg,
+ const uint8_t *buf,
+ VirtioNetRscUnit *n_unit)
+{
+ void *data;
+ uint16_t o_ip_len;
+ uint32_t nseq, oseq;
+ VirtioNetRscUnit *o_unit;
+
+ o_unit = &seg->unit;
+ o_ip_len = htons(*o_unit->ip_plen);
+ nseq = htonl(n_unit->tcp->th_seq);
+ oseq = htonl(o_unit->tcp->th_seq);
+
+ /* out of order or retransmitted. */
+ if ((nseq - oseq) > VIRTIO_NET_MAX_TCP_PAYLOAD) {
+ chain->stat.data_out_of_win++;
+ return RSC_FINAL;
+ }
+
+ data = ((uint8_t *)n_unit->tcp) + n_unit->tcp_hdrlen;
+ if (nseq == oseq) {
+ if ((o_unit->payload == 0) && n_unit->payload) {
+ /* From no payload to payload, normal case, not a dup ack or etc */
+ chain->stat.data_after_pure_ack++;
+ goto coalesce;
+ } else {
+ return virtio_net_rsc_handle_ack(chain, seg, buf,
+ n_unit->tcp, o_unit->tcp);
+ }
+ } else if ((nseq - oseq) != o_unit->payload) {
+ /* Not a consistent packet, out of order */
+ chain->stat.data_out_of_order++;
+ return RSC_FINAL;
+ } else {
+coalesce:
+ if ((o_ip_len + n_unit->payload) > chain->max_payload) {
+ chain->stat.over_size++;
+ return RSC_FINAL;
+ }
+
+ /* Here comes the right data, the payload length in v4/v6 is different,
+ so use the field value to update and record the new data len */
+ o_unit->payload += n_unit->payload; /* update new data len */
+
+ /* update field in ip header */
+ *o_unit->ip_plen = htons(o_ip_len + n_unit->payload);
+
+ /* Bring 'PUSH' big, the whql test guide says 'PUSH' can be coalesced
+ for windows guest, while this may change the behavior for linux
+ guest (only if it uses RSC feature). */
+ o_unit->tcp->th_offset_flags = n_unit->tcp->th_offset_flags;
+
+ o_unit->tcp->th_ack = n_unit->tcp->th_ack;
+ o_unit->tcp->th_win = n_unit->tcp->th_win;
+
+ memmove(seg->buf + seg->size, data, n_unit->payload);
+ seg->size += n_unit->payload;
+ seg->packets++;
+ chain->stat.coalesced++;
+ return RSC_COALESCE;
+ }
+}
+
+static int32_t virtio_net_rsc_coalesce4(VirtioNetRscChain *chain,
+ VirtioNetRscSeg *seg,
+ const uint8_t *buf, size_t size,
+ VirtioNetRscUnit *unit)
+{
+ struct ip_header *ip1, *ip2;
+
+ ip1 = (struct ip_header *)(unit->ip);
+ ip2 = (struct ip_header *)(seg->unit.ip);
+ if ((ip1->ip_src ^ ip2->ip_src) || (ip1->ip_dst ^ ip2->ip_dst)
+ || (unit->tcp->th_sport ^ seg->unit.tcp->th_sport)
+ || (unit->tcp->th_dport ^ seg->unit.tcp->th_dport)) {
+ chain->stat.no_match++;
+ return RSC_NO_MATCH;
+ }
+
+ return virtio_net_rsc_coalesce_data(chain, seg, buf, unit);
+}
+
+static int32_t virtio_net_rsc_coalesce6(VirtioNetRscChain *chain,
+ VirtioNetRscSeg *seg,
+ const uint8_t *buf, size_t size,
+ VirtioNetRscUnit *unit)
+{
+ struct ip6_header *ip1, *ip2;
+
+ ip1 = (struct ip6_header *)(unit->ip);
+ ip2 = (struct ip6_header *)(seg->unit.ip);
+ if (memcmp(&ip1->ip6_src, &ip2->ip6_src, sizeof(struct in6_address))
+ || memcmp(&ip1->ip6_dst, &ip2->ip6_dst, sizeof(struct in6_address))
+ || (unit->tcp->th_sport ^ seg->unit.tcp->th_sport)
+ || (unit->tcp->th_dport ^ seg->unit.tcp->th_dport)) {
+ chain->stat.no_match++;
+ return RSC_NO_MATCH;
+ }
+
+ return virtio_net_rsc_coalesce_data(chain, seg, buf, unit);
+}
+
+/* Packets with 'SYN' should bypass, other flag should be sent after drain
+ * to prevent out of order */
+static int virtio_net_rsc_tcp_ctrl_check(VirtioNetRscChain *chain,
+ struct tcp_header *tcp)
+{
+ uint16_t tcp_hdr;
+ uint16_t tcp_flag;
+
+ tcp_flag = htons(tcp->th_offset_flags);
+ tcp_hdr = (tcp_flag & VIRTIO_NET_TCP_HDR_LENGTH) >> 10;
+ tcp_flag &= VIRTIO_NET_TCP_FLAG;
+ tcp_flag = htons(tcp->th_offset_flags) & 0x3F;
+ if (tcp_flag & TH_SYN) {
+ chain->stat.tcp_syn++;
+ return RSC_BYPASS;
+ }
+
+ if (tcp_flag & (TH_FIN | TH_URG | TH_RST | TH_ECE | TH_CWR)) {
+ chain->stat.tcp_ctrl_drain++;
+ return RSC_FINAL;
+ }
+
+ if (tcp_hdr > sizeof(struct tcp_header)) {
+ chain->stat.tcp_all_opt++;
+ return RSC_FINAL;
+ }
+
+ return RSC_CANDIDATE;
+}
+
+static size_t virtio_net_rsc_do_coalesce(VirtioNetRscChain *chain,
+ NetClientState *nc,
+ const uint8_t *buf, size_t size,
+ VirtioNetRscUnit *unit)
+{
+ int ret;
+ VirtioNetRscSeg *seg, *nseg;
+
+ if (QTAILQ_EMPTY(&chain->buffers)) {
+ chain->stat.empty_cache++;
+ virtio_net_rsc_cache_buf(chain, nc, buf, size);
+ timer_mod(chain->drain_timer,
+ qemu_clock_get_ns(QEMU_CLOCK_HOST) + chain->n->rsc_timeout);
+ return size;
+ }
+
+ QTAILQ_FOREACH_SAFE(seg, &chain->buffers, next, nseg) {
+ if (chain->proto == ETH_P_IP) {
+ ret = virtio_net_rsc_coalesce4(chain, seg, buf, size, unit);
+ } else {
+ ret = virtio_net_rsc_coalesce6(chain, seg, buf, size, unit);
+ }
+
+ if (ret == RSC_FINAL) {
+ if (virtio_net_rsc_drain_seg(chain, seg) == 0) {
+ /* Send failed */
+ chain->stat.final_failed++;
+ return 0;
+ }
+
+ /* Send current packet */
+ return virtio_net_do_receive(nc, buf, size);
+ } else if (ret == RSC_NO_MATCH) {
+ continue;
+ } else {
+ /* Coalesced, mark coalesced flag to tell calc cksum for ipv4 */
+ seg->is_coalesced = 1;
+ return size;
+ }
+ }
+
+ chain->stat.no_match_cache++;
+ virtio_net_rsc_cache_buf(chain, nc, buf, size);
+ return size;
+}
+
+/* Drain a connection data, this is to avoid out of order segments */
+static size_t virtio_net_rsc_drain_flow(VirtioNetRscChain *chain,
+ NetClientState *nc,
+ const uint8_t *buf, size_t size,
+ uint16_t ip_start, uint16_t ip_size,
+ uint16_t tcp_port)
+{
+ VirtioNetRscSeg *seg, *nseg;
+ uint32_t ppair1, ppair2;
+
+ ppair1 = *(uint32_t *)(buf + tcp_port);
+ QTAILQ_FOREACH_SAFE(seg, &chain->buffers, next, nseg) {
+ ppair2 = *(uint32_t *)(seg->buf + tcp_port);
+ if (memcmp(buf + ip_start, seg->buf + ip_start, ip_size)
+ || (ppair1 != ppair2)) {
+ continue;
+ }
+ if (virtio_net_rsc_drain_seg(chain, seg) == 0) {
+ chain->stat.drain_failed++;
+ }
+
+ break;
+ }
+
+ return virtio_net_do_receive(nc, buf, size);
+}
+
+static int32_t virtio_net_rsc_sanity_check4(VirtioNetRscChain *chain,
+ struct ip_header *ip,
+ const uint8_t *buf, size_t size)
+{
+ uint16_t ip_len;
+
+ /* Not an ipv4 packet */
+ if (((ip->ip_ver_len & 0xF0) >> 4) != IP_HEADER_VERSION_4) {
+ chain->stat.ip_option++;
+ return RSC_BYPASS;
+ }
+
+ /* Don't handle packets with ip option */
+ if ((ip->ip_ver_len & 0xF) != VIRTIO_NET_IP4_HEADER_LENGTH) {
+ chain->stat.ip_option++;
+ return RSC_BYPASS;
+ }
+
+ if (ip->ip_p != IPPROTO_TCP) {
+ chain->stat.bypass_not_tcp++;
+ return RSC_BYPASS;
+ }
+
+ /* Don't handle packets with ip fragment */
+ if (!(htons(ip->ip_off) & IP_DF)) {
+ chain->stat.ip_frag++;
+ return RSC_BYPASS;
+ }
+
+ /* Don't handle packets with ecn flag */
+ if (IPTOS_ECN(ip->ip_tos)) {
+ chain->stat.ip_ecn++;
+ return RSC_BYPASS;
+ }
+
+ ip_len = htons(ip->ip_len);
+ if (ip_len < (sizeof(struct ip_header) + sizeof(struct tcp_header))
+ || ip_len > (size - chain->n->guest_hdr_len -
+ sizeof(struct eth_header))) {
+ chain->stat.ip_hacked++;
+ return RSC_BYPASS;
+ }
+
+ return RSC_CANDIDATE;
+}
+
+static size_t virtio_net_rsc_receive4(VirtioNetRscChain *chain,
+ NetClientState *nc,
+ const uint8_t *buf, size_t size)
+{
+ int32_t ret;
+ uint16_t hdr_len;
+ VirtioNetRscUnit unit;
+
+ hdr_len = ((VirtIONet *)(chain->n))->guest_hdr_len;
+
+ if (size < (hdr_len + sizeof(struct eth_header) + sizeof(struct ip_header)
+ + sizeof(struct tcp_header))) {
+ chain->stat.bypass_not_tcp++;
+ return virtio_net_do_receive(nc, buf, size);
+ }
+
+ virtio_net_rsc_extract_unit4(chain, buf, &unit);
+ if (virtio_net_rsc_sanity_check4(chain, unit.ip, buf, size)
+ != RSC_CANDIDATE) {
+ return virtio_net_do_receive(nc, buf, size);
+ }
+
+ ret = virtio_net_rsc_tcp_ctrl_check(chain, unit.tcp);
+ if (ret == RSC_BYPASS) {
+ return virtio_net_do_receive(nc, buf, size);
+ } else if (ret == RSC_FINAL) {
+ return virtio_net_rsc_drain_flow(chain, nc, buf, size,
+ ((hdr_len + sizeof(struct eth_header)) + 12),
+ VIRTIO_NET_IP4_ADDR_SIZE,
+ hdr_len + sizeof(struct eth_header) + sizeof(struct ip_header));
+ }
+
+ return virtio_net_rsc_do_coalesce(chain, nc, buf, size, &unit);
+}
+
+static int32_t virtio_net_rsc_sanity_check6(VirtioNetRscChain *chain,
+ struct ip6_header *ip6,
+ const uint8_t *buf, size_t size)
+{
+ uint16_t ip_len;
+
+ if (((ip6->ip6_ctlun.ip6_un1.ip6_un1_flow & 0xF0) >> 4)
+ != IP_HEADER_VERSION_6) {
+ return RSC_BYPASS;
+ }
+
+ /* Both option and protocol is checked in this */
+ if (ip6->ip6_ctlun.ip6_un1.ip6_un1_nxt != IPPROTO_TCP) {
+ chain->stat.bypass_not_tcp++;
+ return RSC_BYPASS;
+ }
+
+ ip_len = htons(ip6->ip6_ctlun.ip6_un1.ip6_un1_plen);
+ if (ip_len < sizeof(struct tcp_header) ||
+ ip_len > (size - chain->n->guest_hdr_len - sizeof(struct eth_header)
+ - sizeof(struct ip6_header))) {
+ chain->stat.ip_hacked++;
+ return RSC_BYPASS;
+ }
+
+ /* Don't handle packets with ecn flag */
+ if (IP6_ECN(ip6->ip6_ctlun.ip6_un3.ip6_un3_ecn)) {
+ chain->stat.ip_ecn++;
+ return RSC_BYPASS;
+ }
+
+ return RSC_CANDIDATE;
+}
+
+static size_t virtio_net_rsc_receive6(void *opq, NetClientState *nc,
+ const uint8_t *buf, size_t size)
+{
+ int32_t ret;
+ uint16_t hdr_len;
+ VirtioNetRscChain *chain;
+ VirtioNetRscUnit unit;
+
+ chain = (VirtioNetRscChain *)opq;
+ hdr_len = ((VirtIONet *)(chain->n))->guest_hdr_len;
+
+ if (size < (hdr_len + sizeof(struct eth_header) + sizeof(struct ip6_header)
+ + sizeof(tcp_header))) {
+ return virtio_net_do_receive(nc, buf, size);
+ }
+
+ virtio_net_rsc_extract_unit6(chain, buf, &unit);
+ if (RSC_CANDIDATE != virtio_net_rsc_sanity_check6(chain,
+ unit.ip, buf, size)) {
+ return virtio_net_do_receive(nc, buf, size);
+ }
+
+ ret = virtio_net_rsc_tcp_ctrl_check(chain, unit.tcp);
+ if (ret == RSC_BYPASS) {
+ return virtio_net_do_receive(nc, buf, size);
+ } else if (ret == RSC_FINAL) {
+ return virtio_net_rsc_drain_flow(chain, nc, buf, size,
+ ((hdr_len + sizeof(struct eth_header)) + 8),
+ VIRTIO_NET_IP6_ADDR_SIZE,
+ hdr_len + sizeof(struct eth_header)
+ + sizeof(struct ip6_header));
+ }
+
+ return virtio_net_rsc_do_coalesce(chain, nc, buf, size, &unit);
+}
+
+static VirtioNetRscChain *virtio_net_rsc_lookup_chain(VirtIONet *n,
+ NetClientState *nc,
+ uint16_t proto)
+{
+ VirtioNetRscChain *chain;
+
+ if ((proto != (uint16_t)ETH_P_IP) && (proto != (uint16_t)ETH_P_IPV6)) {
+ return NULL;
+ }
+
+ QTAILQ_FOREACH(chain, &n->rsc_chains, next) {
+ if (chain->proto == proto) {
+ return chain;
+ }
+ }
+
+ chain = g_malloc(sizeof(*chain));
+ chain->n = n;
+ chain->proto = proto;
+ if (proto == (uint16_t)ETH_P_IP) {
+ chain->max_payload = VIRTIO_NET_MAX_IP4_PAYLOAD;
+ chain->gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+ } else {
+ chain->max_payload = VIRTIO_NET_MAX_IP6_PAYLOAD;
+ chain->gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+ }
+ chain->drain_timer = timer_new_ns(QEMU_CLOCK_HOST,
+ virtio_net_rsc_purge, chain);
+ memset(&chain->stat, 0, sizeof(chain->stat));
+
+ QTAILQ_INIT(&chain->buffers);
+ QTAILQ_INSERT_TAIL(&n->rsc_chains, chain, next);
+
+ return chain;
+}
+
+static ssize_t virtio_net_rsc_receive(NetClientState *nc,
+ const uint8_t *buf,
+ size_t size)
+{
+ uint16_t proto;
+ VirtioNetRscChain *chain;
+ struct eth_header *eth;
+ VirtIONet *n;
+
+ n = qemu_get_nic_opaque(nc);
+ if (size < (n->host_hdr_len + sizeof(struct eth_header))) {
+ return virtio_net_do_receive(nc, buf, size);
+ }
+
+ eth = (struct eth_header *)(buf + n->guest_hdr_len);
+ proto = htons(eth->h_proto);
+
+ chain = virtio_net_rsc_lookup_chain(n, nc, proto);
+ if (chain) {
+ chain->stat.received++;
+ if (proto == (uint16_t)ETH_P_IP && n->rsc4_enabled) {
+ return virtio_net_rsc_receive4(chain, nc, buf, size);
+ } else if (proto == (uint16_t)ETH_P_IPV6 && n->rsc6_enabled) {
+ return virtio_net_rsc_receive6(chain, nc, buf, size);
+ }
+ }
+ return virtio_net_do_receive(nc, buf, size);
+}
+
+static ssize_t virtio_net_receive(NetClientState *nc, const uint8_t *buf,
+ size_t size)
+{
+ VirtIONet *n = qemu_get_nic_opaque(nc);
+ if ((n->rsc4_enabled || n->rsc6_enabled)) {
+ return virtio_net_rsc_receive(nc, buf, size);
+ } else {
+ return virtio_net_do_receive(nc, buf, size);
+ }
+}
+
static int32_t virtio_net_flush_tx(VirtIONetQueue *q);
static void virtio_net_tx_complete(NetClientState *nc, ssize_t len)
@@ -2075,6 +2734,7 @@ static void virtio_net_device_realize(DeviceState *dev, Error **errp)
nc = qemu_get_queue(n->nic);
nc->rxfilter_notify_enabled = 1;
+ QTAILQ_INIT(&n->rsc_chains);
n->qdev = dev;
}
@@ -2104,6 +2764,7 @@ static void virtio_net_device_unrealize(DeviceState *dev, Error **errp)
timer_free(n->announce_timer);
g_free(n->vqs);
qemu_del_nic(n->nic);
+ virtio_net_rsc_cleanup(n);
virtio_cleanup(vdev);
}
@@ -2184,6 +2845,10 @@ static Property virtio_net_properties[] = {
DEFINE_PROP_BIT64("ctrl_guest_offloads", VirtIONet, host_features,
VIRTIO_NET_F_CTRL_GUEST_OFFLOADS, true),
DEFINE_PROP_BIT64("mq", VirtIONet, host_features, VIRTIO_NET_F_MQ, false),
+ DEFINE_PROP_BIT64("guest_rsc_ext", VirtIONet, host_features,
+ VIRTIO_NET_F_RSC_EXT, false),
+ DEFINE_PROP_UINT32("rsc_interval", VirtIONet, rsc_timeout,
+ VIRTIO_NET_RSC_DEFAULT_INTERVAL),
DEFINE_NIC_PROPERTIES(VirtIONet, nic_conf),
DEFINE_PROP_UINT32("x-txtimer", VirtIONet, net_conf.txtimer,
TX_TIMER_INTERVAL),
@@ -44,6 +44,83 @@ typedef struct virtio_net_conf
uint8_t duplex;
} virtio_net_conf;
+/* Coalesced packets type & status */
+typedef enum {
+ RSC_COALESCE, /* Data been coalesced */
+ RSC_FINAL, /* Will terminate current connection */
+ RSC_NO_MATCH, /* No matched in the buffer pool */
+ RSC_BYPASS, /* Packet to be bypass, not tcp, tcp ctrl, etc */
+ RSC_CANDIDATE /* Data want to be coalesced */
+} CoalesceStatus;
+
+typedef struct VirtioNetRscStat {
+ uint32_t received;
+ uint32_t coalesced;
+ uint32_t over_size;
+ uint32_t cache;
+ uint32_t empty_cache;
+ uint32_t no_match_cache;
+ uint32_t win_update;
+ uint32_t no_match;
+ uint32_t tcp_syn;
+ uint32_t tcp_ctrl_drain;
+ uint32_t dup_ack;
+ uint32_t dup_ack1;
+ uint32_t dup_ack2;
+ uint32_t pure_ack;
+ uint32_t ack_out_of_win;
+ uint32_t data_out_of_win;
+ uint32_t data_out_of_order;
+ uint32_t data_after_pure_ack;
+ uint32_t bypass_not_tcp;
+ uint32_t tcp_option;
+ uint32_t tcp_all_opt;
+ uint32_t ip_frag;
+ uint32_t ip_ecn;
+ uint32_t ip_hacked;
+ uint32_t ip_option;
+ uint32_t purge_failed;
+ uint32_t drain_failed;
+ uint32_t final_failed;
+ int64_t timer;
+} VirtioNetRscStat;
+
+/* Rsc unit general info used to checking if can coalescing */
+typedef struct VirtioNetRscUnit {
+ void *ip; /* ip header */
+ uint16_t *ip_plen; /* data len pointer in ip header field */
+ struct tcp_header *tcp; /* tcp header */
+ uint16_t tcp_hdrlen; /* tcp header len */
+ uint16_t payload; /* pure payload without virtio/eth/ip/tcp */
+} VirtioNetRscUnit;
+
+/* Coalesced segmant */
+typedef struct VirtioNetRscSeg {
+ QTAILQ_ENTRY(VirtioNetRscSeg) next;
+ void *buf;
+ size_t size;
+ uint16_t packets;
+ uint16_t dup_ack;
+ bool is_coalesced; /* need recal ipv4 header checksum, mark here */
+ VirtioNetRscUnit unit;
+ NetClientState *nc;
+} VirtioNetRscSeg;
+
+struct VirtIONet;
+typedef struct VirtIONet VirtIONet;
+
+/* Chain is divided by protocol(ipv4/v6) and NetClientInfo */
+typedef struct VirtioNetRscChain {
+ QTAILQ_ENTRY(VirtioNetRscChain) next;
+ VirtIONet *n; /* VirtIONet */
+ uint16_t proto;
+ uint8_t gso_type;
+ uint16_t max_payload;
+ QEMUTimer *drain_timer;
+ QTAILQ_HEAD(, VirtioNetRscSeg) buffers;
+ VirtioNetRscStat stat;
+} VirtioNetRscChain;
+
/* Maximum packet size we can receive from tap device: header + 64k */
#define VIRTIO_NET_MAX_BUFSIZE (sizeof(struct virtio_net_hdr) + (64 * KiB))
@@ -66,12 +143,18 @@ typedef struct VirtIONet {
VirtIONetQueue *vqs;
VirtQueue *ctrl_vq;
NICState *nic;
+ /* RSC Chains - temporary storage of coalesced data,
+ all these data are lost in case of migration */
+ QTAILQ_HEAD(, VirtioNetRscChain) rsc_chains;
uint32_t tx_timeout;
int32_t tx_burst;
uint32_t has_vnet_hdr;
size_t host_hdr_len;
size_t guest_hdr_len;
uint64_t host_features;
+ uint32_t rsc_timeout;
+ uint8_t rsc4_enabled;
+ uint8_t rsc6_enabled;
uint8_t has_ufo;
uint32_t mergeable_rx_bufs;
uint8_t promisc;
@@ -177,6 +177,8 @@ struct tcp_hdr {
#define TH_PUSH 0x08
#define TH_ACK 0x10
#define TH_URG 0x20
+#define TH_ECE 0x40
+#define TH_CWR 0x80
u_short th_win; /* window */
u_short th_sum; /* checksum */
u_short th_urp; /* urgent pointer */