Re: [PATCH 3/4] multiqueue: Add queue-aware flow management for multiqueue support

On Fri, Nov 07, 2025 at 03:39:00PM +0100, Laurent Vivier wrote:

Packets are now routed to the correct RX queue based on which TX queue they arrived on, rather than always using queue 0.

Am I missing something, or is this not quite accurate. The packets are associated by flow with other packets that came on a particular Tx queue, but AIUI the packet itself didn't arrive on a Tx queue, but rather from a socket.

Note: Flows initiated from the host (via sockets, udp_flow_from_sock()) currently default to queue 0, as they don't have an associated incoming queue.

Signed-off-by: Laurent Vivier --- flow.c | 32 ++++++++++++++++++++++ flow.h | 10 +++++++ icmp.c | 23 +++++++++------- icmp.h | 2 +- tap.c | 77 +++++++++++++++++++++++++++------------------------ tap.h | 5 ++-- tcp.c | 79 +++++++++++++++++++++++++++++------------------------ tcp.h | 2 +- tcp_vu.c | 8 ++++-- udp.c | 33 ++++++++++++---------- udp.h | 12 ++++---- udp_flow.c | 8 +++++- udp_flow.h | 2 +- udp_vu.c | 4 ++- vu_common.c | 4 +-- 15 files changed, 187 insertions(+), 114 deletions(-)

diff --git a/flow.c b/flow.c index 278a9cf0ac6d..8e9d7e5e1847 100644 --- a/flow.c +++ b/flow.c @@ -405,6 +405,37 @@ void flow_epollid_register(int epollid, int epollfd) epoll_id_to_fd[epollid] = epollfd; }

+/** + * flow_rx_virtqueue() - Get RX (receive) queue number for a flow

Maybe avoid the name "virtqueue". I believe the tap device can also do multiqueue, and we may want to reuse this queue infrastructure for that.

+ * @f: Flow to query (may be NULL) + * + * Return: RX queue number for the flow, or 0 if flow is NULL or has no + * valid queue assignment + */ +int flow_rx_virtqueue(const struct flow_common *f) +{ + if (f == NULL || f->queueid == FLOW_QUEUEID_INVALID)

Any reason to have the second special case here, rather than initializing f->queueid to 0?

+ return 0; + return f->queueid << 1;

I generally think it's clearer to use * 2 rather than << 1 here. I expect the compiler to turn it into the same thing.

+} + +/** + * flow_queue_set() - Set queue pair assignment for a flow + * @f: Flow to update + * @queueid: Queue pair number to assign (even number, RX queue; TX is RX+1)

In comments using "from guest" / "to guest" instead of Rx/Tx is probably a good idea. Particularly in code that's not necessarily specific to VU. I think it is a bit confusing that somewhere we have queue pair IDs, sometimes absolute queue numbers. I think we should standardise on one throughout as much of the code as possible.

+ */ +void flow_queue_set(struct flow_common *f, int queueid) +{ + queueid >>= 1; + + ASSERT(queueid < FLOW_QUEUEID_MAX); + + flow_trace((union flow *)f, "updating queue from %d to %d", + f->queueid, queueid); + + f->queueid = queueid; +} + /** * flow_initiate_() - Move flow to INI, setting pif[INISIDE] * @flow: Flow to change state @@ -609,6 +640,7 @@ union flow *flow_alloc(void) flow_new_entry = flow; memset(flow, 0, sizeof(*flow)); flow_epollid_clear(&flow->f); + flow->f.queueid = FLOW_QUEUEID_INVALID; flow_set_state(&flow->f, FLOW_STATE_NEW);

return flow; diff --git a/flow.h b/flow.h index b43b0b1dd7f2..44ab4ae8fd6a 100644 --- a/flow.h +++ b/flow.h @@ -179,6 +179,8 @@ int flowside_connect(const struct ctx *c, int s, * @side[]: Information for each side of the flow * @tap_omac: MAC address of remote endpoint as seen from the guest * @epollid: epollfd identifier, or EPOLLFD_ID_INVALID + * @queueid: Queue pair number assigned to this flow + * (FLOW_QUEUEID_INVALID if not assigned) */ struct flow_common { #ifdef __GNUC__ @@ -199,6 +201,8 @@ struct flow_common {

#define EPOLLFD_ID_BITS 8 unsigned int epollid:EPOLLFD_ID_BITS; +#define FLOW_QUEUEID_BITS 5 + unsigned int queueid:FLOW_QUEUEID_BITS; };

#define EPOLLFD_ID_DEFAULT 0 @@ -206,6 +210,10 @@ struct flow_common { #define EPOLLFD_ID_MAX (EPOLLFD_ID_SIZE - 1) #define EPOLLFD_ID_INVALID EPOLLFD_ID_MAX

+#define FLOW_QUEUEID_SIZE (1 << FLOW_QUEUEID_BITS)

"SIZE" is a bit ambiguous here. Maybe "NUM" instead?

+#define FLOW_QUEUEID_MAX (FLOW_QUEUEID_SIZE - 1) +#define FLOW_QUEUEID_INVALID FLOW_QUEUEID_MAX + #define FLOW_INDEX_BITS 17 /* 128k - 1 */ #define FLOW_MAX MAX_FROM_BITS(FLOW_INDEX_BITS)

@@ -266,6 +274,8 @@ int flow_epollfd(const struct flow_common *f); void flow_epollid_set(struct flow_common *f, int epollid); void flow_epollid_clear(struct flow_common *f); void flow_epollid_register(int epollid, int epollfd); +int flow_rx_virtqueue(const struct flow_common *f); +void flow_queue_set(struct flow_common *f, int queueid); void flow_defer_handler(const struct ctx *c, const struct timespec *now); int flow_migrate_source_early(struct ctx *c, const struct migrate_stage *stage, int fd); diff --git a/icmp.c b/icmp.c index d58499c3bf5c..80e8753072fa 100644 --- a/icmp.c +++ b/icmp.c @@ -132,13 +132,13 @@ void icmp_sock_handler(const struct ctx *c, union epoll_ref ref) const struct in_addr *daddr = inany_v4(&ini->eaddr);

ASSERT(saddr && daddr); /* Must have IPv4 addresses */ - tap_icmp4_send(c, VHOST_USER_RX_QUEUE, *saddr, *daddr, buf, + tap_icmp4_send(c, flow_rx_virtqueue(&pingf->f), *saddr, *daddr, buf, pingf->f.tap_omac, n); } else if (pingf->f.type == FLOW_PING6) { const struct in6_addr *saddr = &ini->oaddr.a6; const struct in6_addr *daddr = &ini->eaddr.a6;

- tap_icmp6_send(c, VHOST_USER_RX_QUEUE, saddr, daddr, buf, + tap_icmp6_send(c, flow_rx_virtqueue(&pingf->f), saddr, daddr, buf, pingf->f.tap_omac, n); } return; @@ -238,17 +238,18 @@ cancel:

/** * icmp_tap_handler() - Handle packets from tap - * @c: Execution context - * @pif: pif on which the packet is arriving - * @af: Address family, AF_INET or AF_INET6 - * @saddr: Source address - * @daddr: Destination address - * @data: Single packet with ICMP/ICMPv6 header - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @pif: pif on which the packet is arriving + * @af: Address family, AF_INET or AF_INET6 + * @saddr: Source address + * @daddr: Destination address + * @data: Single packet with ICMP/ICMPv6 header + * @now: Current timestamp * * Return: count of consumed packets (always 1, even if malformed) */ -int icmp_tap_handler(const struct ctx *c, uint8_t pif, sa_family_t af, +int icmp_tap_handler(const struct ctx *c, int incoming_queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, struct iov_tail *data, const struct timespec *now) { @@ -309,6 +310,8 @@ int icmp_tap_handler(const struct ctx *c, uint8_t pif, sa_family_t af, else if (!(pingf = icmp_ping_new(c, af, id, saddr, daddr))) return 1;

+ flow_queue_set(&pingf->f, incoming_queue); +

I'd kind of like the initial setting of the flow's queue to be tied to one of the existing flow state transitions, to ensure it can't be forgotten.

tgt = &pingf->f.side[TGTSIDE];

ASSERT(flow_proto[pingf->f.type] == proto); diff --git a/icmp.h b/icmp.h index 1a0e6205f087..6d6d6358bb33 100644 --- a/icmp.h +++ b/icmp.h @@ -10,7 +10,7 @@ struct ctx; struct icmp_ping_flow;

void icmp_sock_handler(const struct ctx *c, union epoll_ref ref); -int icmp_tap_handler(const struct ctx *c, uint8_t pif, sa_family_t af, +int icmp_tap_handler(const struct ctx *c, int incoming_queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, struct iov_tail *data, const struct timespec *now); void icmp_init(void); diff --git a/tap.c b/tap.c index 1308d49242e8..9a4399d947a3 100644 --- a/tap.c +++ b/tap.c @@ -702,15 +702,17 @@ static bool tap4_is_fragment(const struct iphdr *iph,

/** * tap4_handler() - IPv4 and ARP packet handler for tap file descriptor - * @c: Execution context - * @in: Ingress packet pool, packets with Ethernet headers - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @in: Ingress packet pool, packets with Ethernet headers + * @now: Current timestamp * * Return: count of packets consumed by handlers */ -static int tap4_handler(struct ctx *c, const struct pool *in, - const struct timespec *now) +static int tap4_handler(struct ctx *c, int incoming_queue, + const struct pool *in, const struct timespec *now) { + int outgoing_queue = incoming_queue & ~1; unsigned int i, j, seq_count; struct tap4_l4_t *seq;

@@ -736,7 +738,7 @@ resume: if (!eh) continue; if (ntohs(eh->h_proto) == ETH_P_ARP) { - arp(c, VHOST_USER_RX_QUEUE, &data); + arp(c, outgoing_queue, &data); continue; }

@@ -783,7 +785,7 @@ resume:

tap_packet_debug(iph, NULL, NULL, 0, NULL, 1);

- icmp_tap_handler(c, PIF_TAP, AF_INET, + icmp_tap_handler(c, incoming_queue, PIF_TAP, AF_INET, &iph->saddr, &iph->daddr, &data, now); continue; @@ -797,7 +799,7 @@ resume: struct iov_tail eh_data;

packet_get(in, i, &eh_data); - if (dhcp(c, VHOST_USER_RX_QUEUE, &eh_data)) + if (dhcp(c, outgoing_queue, &eh_data)) continue; }

@@ -860,14 +862,14 @@ append: if (c->no_tcp) continue; for (k = 0; k < p->count; ) - k += tcp_tap_handler(c, PIF_TAP, AF_INET, + k += tcp_tap_handler(c, incoming_queue, PIF_TAP, AF_INET, &seq->saddr, &seq->daddr, 0, p, k, now); } else if (seq->protocol == IPPROTO_UDP) { if (c->no_udp) continue; for (k = 0; k < p->count; ) - k += udp_tap_handler(c, PIF_TAP, AF_INET, + k += udp_tap_handler(c, incoming_queue, PIF_TAP, AF_INET, &seq->saddr, &seq->daddr, seq->ttl, p, k, now); } @@ -881,15 +883,17 @@ append:

/** * tap6_handler() - IPv6 packet handler for tap file descriptor - * @c: Execution context - * @in: Ingress packet pool, packets with Ethernet headers - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @in: Ingress packet pool, packets with Ethernet headers + * @now: Current timestamp * * Return: count of packets consumed by handlers */ -static int tap6_handler(struct ctx *c, const struct pool *in, - const struct timespec *now) +static int tap6_handler(struct ctx *c, int incoming_queue, + const struct pool *in, const struct timespec *now) { + int outgoing_queue = incoming_queue & ~1; unsigned int i, j, seq_count = 0; struct tap6_l4_t *seq;

@@ -965,12 +969,12 @@ resume: continue;

ndp_data = data; - if (ndp(c, VHOST_USER_RX_QUEUE, saddr, &ndp_data)) + if (ndp(c, outgoing_queue, saddr, &ndp_data)) continue;

tap_packet_debug(NULL, ip6h, NULL, proto, NULL, 1);

- icmp_tap_handler(c, PIF_TAP, AF_INET6, + icmp_tap_handler(c, incoming_queue, PIF_TAP, AF_INET6, saddr, daddr, &data, now); continue; } @@ -984,8 +988,7 @@ resume: if (proto == IPPROTO_UDP) { struct iov_tail uh_data = data;

- if (dhcpv6(c, VHOST_USER_RX_QUEUE, &uh_data, saddr, - daddr)) + if (dhcpv6(c, outgoing_queue, &uh_data, saddr, daddr)) continue; }

@@ -1053,14 +1056,14 @@ append: if (c->no_tcp) continue; for (k = 0; k < p->count; ) - k += tcp_tap_handler(c, PIF_TAP, AF_INET6, + k += tcp_tap_handler(c, incoming_queue, PIF_TAP, AF_INET6, &seq->saddr, &seq->daddr, seq->flow_lbl, p, k, now); } else if (seq->protocol == IPPROTO_UDP) { if (c->no_udp) continue; for (k = 0; k < p->count; ) - k += udp_tap_handler(c, PIF_TAP, AF_INET6, + k += udp_tap_handler(c, incoming_queue, PIF_TAP, AF_INET6, &seq->saddr, &seq->daddr, seq->hop_limit, p, k, now); } @@ -1083,22 +1086,24 @@ void tap_flush_pools(void)

/** * tap_handler() - IPv4/IPv6 and ARP packet handler for tap file descriptor - * @c: Execution context - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @now: Current timestamp */ -void tap_handler(struct ctx *c, const struct timespec *now) +void tap_handler(struct ctx *c, int incoming_queue, const struct timespec *now) { - tap4_handler(c, pool_tap4, now); - tap6_handler(c, pool_tap6, now); + tap4_handler(c, incoming_queue, pool_tap4, now); + tap6_handler(c, incoming_queue, pool_tap6, now); }

/** * tap_add_packet() - Queue/capture packet, update notion of guest MAC address - * @c: Execution context - * @data: Packet to add to the pool - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @data: Packet to add to the pool + * @now: Current timestamp */ -void tap_add_packet(struct ctx *c, struct iov_tail *data, +void tap_add_packet(struct ctx *c, int incoming_queue, struct iov_tail *data, const struct timespec *now) { struct ethhdr eh_storage; @@ -1123,14 +1128,14 @@ void tap_add_packet(struct ctx *c, struct iov_tail *data, case ETH_P_ARP: case ETH_P_IP: if (!pool_can_fit(pool_tap4, data)) { - tap4_handler(c, pool_tap4, now); + tap4_handler(c, incoming_queue, pool_tap4, now);

This is a bit murky. You're using the incoming_queue for the new packet to process all the existiing packets in the pool. I guess that will work out because you'll have a separate pool for each thread/queue. But that's kind of unclear at this point. Maybe the queue number should be more explicitly part of the pool metadata?

pool_flush(pool_tap4); } packet_add(pool_tap4, data); break; case ETH_P_IPV6: if (!pool_can_fit(pool_tap6, data)) { - tap6_handler(c, pool_tap6, now); + tap6_handler(c, incoming_queue, pool_tap6, now); pool_flush(pool_tap6); } packet_add(pool_tap6, data); @@ -1217,7 +1222,7 @@ static void tap_passt_input(struct ctx *c, const struct timespec *now) n -= sizeof(uint32_t);

data = IOV_TAIL_FROM_BUF(p, l2len, 0); - tap_add_packet(c, &data, now); + tap_add_packet(c, 0, &data, now);

p += l2len; n -= l2len; @@ -1226,7 +1231,7 @@ static void tap_passt_input(struct ctx *c, const struct timespec *now) partial_len = n; partial_frame = p;

- tap_handler(c, now); + tap_handler(c, 0, now); }

/** @@ -1285,10 +1290,10 @@ static void tap_pasta_input(struct ctx *c, const struct timespec *now) continue;

data = IOV_TAIL_FROM_BUF(pkt_buf + n, len, 0); - tap_add_packet(c, &data, now); + tap_add_packet(c, 0, &data, now); }

- tap_handler(c, now); + tap_handler(c, 0, now); }

/** diff --git a/tap.h b/tap.h index 76403a43edbc..fe9455ffcf4b 100644 --- a/tap.h +++ b/tap.h @@ -119,7 +119,8 @@ int tap_sock_unix_open(char *sock_path); void tap_sock_reset(struct ctx *c); void tap_backend_init(struct ctx *c); void tap_flush_pools(void); -void tap_handler(struct ctx *c, const struct timespec *now); -void tap_add_packet(struct ctx *c, struct iov_tail *data, +void tap_handler(struct ctx *c, int incoming_queue, + const struct timespec *now); +void tap_add_packet(struct ctx *c, int incoming_queue, struct iov_tail *data, const struct timespec *now); #endif /* TAP_H */ diff --git a/tcp.c b/tcp.c index 5ce34baa8a5a..78494a2dc69f 100644 --- a/tcp.c +++ b/tcp.c @@ -1496,21 +1496,23 @@ static void tcp_bind_outbound(const struct ctx *c,

/** * tcp_conn_from_tap() - Handle connection request (SYN segment) from tap - * @c: Execution context - * @af: Address family, AF_INET or AF_INET6 - * @saddr: Source address, pointer to in_addr or in6_addr - * @daddr: Destination address, pointer to in_addr or in6_addr - * @th: TCP header from tap: caller MUST ensure it's there - * @opts: Pointer to start of options - * @optlen: Bytes in options: caller MUST ensure available length - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: TX queue number for the flow + * @af: Address family, AF_INET or AF_INET6 + * @saddr: Source address, pointer to in_addr or in6_addr + * @daddr: Destination address, pointer to in_addr or in6_addr + * @th: TCP header from tap: caller MUST ensure it's there + * @opts: Pointer to start of options + * @optlen: Bytes in options: caller MUST ensure available length + * @now: Current timestamp * * #syscalls:vu getsockname */ -static void tcp_conn_from_tap(const struct ctx *c, sa_family_t af, - const void *saddr, const void *daddr, - const struct tcphdr *th, const char *opts, - size_t optlen, const struct timespec *now) +static void tcp_conn_from_tap(const struct ctx *c, int incoming_queue, + sa_family_t af, const void *saddr, + const void *daddr, const struct tcphdr *th, + const char *opts, size_t optlen, + const struct timespec *now) { in_port_t srcport = ntohs(th->source); in_port_t dstport = ntohs(th->dest); @@ -1622,6 +1624,7 @@ static void tcp_conn_from_tap(const struct ctx *c, sa_family_t af, conn_event(c, conn, TAP_SYN_ACK_SENT); }

+ flow_queue_set(&conn->f, incoming_queue); tcp_epoll_ctl(c, conn);

if (c->mode == MODE_VU) { /* To rebind to same oport after migration */ @@ -1983,16 +1986,16 @@ static void tcp_conn_from_sock_finish(const struct ctx *c,

/** * tcp_rst_no_conn() - Send RST in response to a packet with no connection - * @c: Execution context - * @queue: Queue to use to send the reply - * @af: Address family, AF_INET or AF_INET6 - * @saddr: Source address of the packet we're responding to - * @daddr: Destination address of the packet we're responding to - * @flow_lbl: IPv6 flow label (ignored for IPv4) - * @th: TCP header of the packet we're responding to - * @l4len: Packet length, including TCP header - */ -static void tcp_rst_no_conn(const struct ctx *c, int queue, int af, + * @c: Execution context + * @outgoing_queue: Queue to use to send the reply + * @af: Address family, AF_INET or AF_INET6 + * @saddr: Source address of the packet we're responding to + * @daddr: Destination address of the packet we're responding to + * @flow_lbl: IPv6 flow label (ignored for IPv4) + * @th: TCP header of the packet we're responding to + * @l4len: Packet length, including TCP header + */ +static void tcp_rst_no_conn(const struct ctx *c, int outgoing_queue, int af, const void *saddr, const void *daddr, uint32_t flow_lbl, const struct tcphdr *th, size_t l4len) @@ -2050,24 +2053,25 @@ static void tcp_rst_no_conn(const struct ctx *c, int queue, int af,

tcp_update_csum(psum, rsth, &payload); rst_l2len = ((char *)rsth - buf) + sizeof(*rsth); - tap_send_single(c, queue, buf, rst_l2len); + tap_send_single(c, outgoing_queue, buf, rst_l2len); }

/** * tcp_tap_handler() - Handle packets from tap and state transitions - * @c: Execution context - * @pif: pif on which the packet is arriving - * @af: Address family, AF_INET or AF_INET6 - * @saddr: Source address - * @daddr: Destination address - * @flow_lbl: IPv6 flow label (ignored for IPv4) - * @p: Pool of TCP packets, with TCP headers - * @idx: Index of first packet in pool to process - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @pif: pif on which the packet is arriving + * @af: Address family, AF_INET or AF_INET6 + * @saddr: Source address + * @daddr: Destination address + * @flow_lbl: IPv6 flow label (ignored for IPv4) + * @p: Pool of TCP packets, with TCP headers + * @idx: Index of first packet in pool to process + * @now: Current timestamp * * Return: count of consumed packets */ -int tcp_tap_handler(const struct ctx *c, uint8_t pif, +int tcp_tap_handler(const struct ctx *c, int incoming_queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, uint32_t flow_lbl, const struct pool *p, int idx, const struct timespec *now) @@ -2107,11 +2111,11 @@ int tcp_tap_handler(const struct ctx *c, uint8_t pif, /* New connection from tap */ if (!flow) { if (opts && th->syn && !th->ack) - tcp_conn_from_tap(c, af, saddr, daddr, th, + tcp_conn_from_tap(c, incoming_queue, af, saddr, daddr, th, opts, optlen, now); else - tcp_rst_no_conn(c, VHOST_USER_RX_QUEUE, af, saddr, - daddr, flow_lbl, th, l4len); + tcp_rst_no_conn(c, incoming_queue & ~1, af, saddr, daddr, + flow_lbl, th, l4len); return 1; }

@@ -2119,6 +2123,9 @@ int tcp_tap_handler(const struct ctx *c, uint8_t pif, ASSERT(pif_at_sidx(sidx) == PIF_TAP); conn = &flow->tcp;

+ /* update queue */ + flow_queue_set(&flow->f, incoming_queue); + flow_trace(conn, "packet length %zu from tap", l4len);

if (th->rst) { diff --git a/tcp.h b/tcp.h index 320683ce5679..cddd36cadc97 100644 --- a/tcp.h +++ b/tcp.h @@ -15,7 +15,7 @@ void tcp_listen_handler(const struct ctx *c, union epoll_ref ref, const struct timespec *now); void tcp_sock_handler(const struct ctx *c, union epoll_ref ref, uint32_t events); -int tcp_tap_handler(const struct ctx *c, uint8_t pif, +int tcp_tap_handler(const struct ctx *c, int incoming_queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, uint32_t flow_lbl, const struct pool *p, int idx, const struct timespec *now); diff --git a/tcp_vu.c b/tcp_vu.c index 1c81ce376dad..40f552087bc5 100644 --- a/tcp_vu.c +++ b/tcp_vu.c @@ -70,15 +70,16 @@ static size_t tcp_vu_hdrlen(bool v6) */ int tcp_vu_send_flag(const struct ctx *c, struct tcp_tap_conn *conn, int flags) { + int rx_queue = flow_rx_virtqueue(&conn->f); struct vu_dev *vdev = c->vdev; - struct vu_virtq *vq = &vdev->vq[VHOST_USER_RX_QUEUE]; - size_t optlen, hdrlen; + struct vu_virtq *vq = &vdev->vq[rx_queue]; struct vu_virtq_element flags_elem[2]; struct ipv6hdr *ip6h = NULL; struct iphdr *ip4h = NULL; struct iovec flags_iov[2]; struct tcp_syn_opts *opts; struct iov_tail payload; + size_t optlen, hdrlen; struct tcphdr *th; struct ethhdr *eh; uint32_t seq; @@ -348,8 +349,9 @@ static void tcp_vu_prepare(const struct ctx *c, struct tcp_tap_conn *conn, int tcp_vu_data_from_sock(const struct ctx *c, struct tcp_tap_conn *conn) { uint32_t wnd_scaled = conn->wnd_from_tap << conn->ws_from_tap; + int rx_queue = flow_rx_virtqueue(&conn->f); struct vu_dev *vdev = c->vdev; - struct vu_virtq *vq = &vdev->vq[VHOST_USER_RX_QUEUE]; + struct vu_virtq *vq = &vdev->vq[rx_queue]; ssize_t len, previous_dlen; int i, iov_cnt, head_cnt; size_t hdrlen, fillsize; diff --git a/udp.c b/udp.c index 868ffebb5802..a0eb719888cd 100644 --- a/udp.c +++ b/udp.c @@ -636,14 +636,15 @@ static int udp_sock_recverr(const struct ctx *c, int s, flow_sidx_t sidx, if (hdr->cmsg_level == IPPROTO_IP && (o4 = inany_v4(&otap)) && inany_v4(&toside->eaddr)) { dlen = MIN(dlen, ICMP4_MAX_DLEN); - udp_send_tap_icmp4(c, VHOST_USER_RX_QUEUE, ee, toside, *o4, - data, dlen); + udp_send_tap_icmp4(c, flow_rx_virtqueue(&uflow->f), ee, toside,

Kind of a something for an earlier patch, but given how frequently and where it's used, it's worth having a shorter name for flow_rx_virtqueue(). And probably a wrapper macro so it can be called on uflow directly as well.

+ *o4, data, dlen); return 1; }

if (hdr->cmsg_level == IPPROTO_IPV6 && !inany_v4(&toside->eaddr)) { - udp_send_tap_icmp6(c, VHOST_USER_RX_QUEUE, ee, toside, - &otap.a6, data, dlen, FLOW_IDX(uflow)); + udp_send_tap_icmp6(c, flow_rx_virtqueue(&uflow->f), ee, + toside, &otap.a6, data, dlen, + FLOW_IDX(uflow)); return 1; }

@@ -971,25 +972,27 @@ fail:

/** * udp_tap_handler() - Handle packets from tap - * @c: Execution context - * @pif: pif on which the packet is arriving - * @af: Address family, AF_INET or AF_INET6 - * @saddr: Source address - * @daddr: Destination address - * @ttl: TTL or hop limit for packets to be sent in this call - * @p: Pool of UDP packets, with UDP headers - * @idx: Index of first packet to process - * @now: Current timestamp + * @c: Execution context + * @incoming_queue: Incoming queue number + * @pif: pif on which the packet is arriving + * @af: Address family, AF_INET or AF_INET6 + * @saddr: Source address + * @daddr: Destination address + * @ttl: TTL or hop limit for packets to be sent in this call + * @p: Pool of UDP packets, with UDP headers + * @idx: Index of first packet to process + * @now: Current timestamp * * Return: count of consumed packets * * #syscalls sendmmsg */ -int udp_tap_handler(const struct ctx *c, uint8_t pif, +int udp_tap_handler(const struct ctx *c, int incoming_queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, uint8_t ttl, const struct pool *p, int idx, const struct timespec *now) { + int outgoing_queue = incoming_queue & ~1; const struct flowside *toside; struct mmsghdr mm[UIO_MAXIOV]; union sockaddr_inany to_sa; @@ -1019,7 +1022,7 @@ int udp_tap_handler(const struct ctx *c, uint8_t pif, src = ntohs(uh->source); dst = ntohs(uh->dest);

- tosidx = udp_flow_from_tap(c, pif, af, saddr, daddr, src, dst, now); + tosidx = udp_flow_from_tap(c, outgoing_queue, pif, af, saddr, daddr, src, dst, now); if (!(uflow = udp_at_sidx(tosidx))) { char sstr[INET6_ADDRSTRLEN], dstr[INET6_ADDRSTRLEN];

diff --git a/udp.h b/udp.h index f1d83f380b3f..8ba4ccfe646a 100644 --- a/udp.h +++ b/udp.h @@ -7,11 +7,13 @@ #define UDP_H

void udp_portmap_clear(void); -void udp_listen_sock_handler(const struct ctx *c, union epoll_ref ref, - uint32_t events, const struct timespec *now); -void udp_sock_handler(const struct ctx *c, union epoll_ref ref, - uint32_t events, const struct timespec *now); -int udp_tap_handler(const struct ctx *c, uint8_t pif, +void udp_listen_sock_handler(const struct ctx *c, + union epoll_ref ref, uint32_t events, + const struct timespec *now); +void udp_sock_handler(const struct ctx *c, + union epoll_ref ref, uint32_t events, + const struct timespec *now); +int udp_tap_handler(const struct ctx *c, int incoming_queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, uint8_t ttl, const struct pool *p, int idx, const struct timespec *now); diff --git a/udp_flow.c b/udp_flow.c index 8907f2f72741..b4a709b8d976 100644 --- a/udp_flow.c +++ b/udp_flow.c @@ -266,17 +266,19 @@ flow_sidx_t udp_flow_from_sock(const struct ctx *c, uint8_t pif, /** * udp_flow_from_tap() - Find or create UDP flow for tap packets * @c: Execution context + * @queue: RX queue number for the flow * @pif: pif on which the packet is arriving * @af: Address family, AF_INET or AF_INET6 * @saddr: Source address on guest side * @daddr: Destination address guest side * @srcport: Source port on guest side * @dstport: Destination port on guest side + * @now: Current timestamp * * Return: sidx for the destination side of the flow for this packet, or * FLOW_SIDX_NONE if we couldn't find or create a flow. */ -flow_sidx_t udp_flow_from_tap(const struct ctx *c, +flow_sidx_t udp_flow_from_tap(const struct ctx *c, int queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, in_port_t srcport, in_port_t dstport, @@ -293,6 +295,8 @@ flow_sidx_t udp_flow_from_tap(const struct ctx *c, srcport, dstport); if ((uflow = udp_at_sidx(sidx))) { uflow->ts = now->tv_sec; + /* update queue */ + flow_queue_set(&uflow->f, queue); return flow_sidx_opposite(sidx); }

@@ -316,6 +320,8 @@ flow_sidx_t udp_flow_from_tap(const struct ctx *c, return FLOW_SIDX_NONE; }

+ flow_queue_set(&flow->f, queue); + return udp_flow_new(c, flow, now); }

diff --git a/udp_flow.h b/udp_flow.h index 4c528e95ca66..4a057a9d44a8 100644 --- a/udp_flow.h +++ b/udp_flow.h @@ -36,7 +36,7 @@ flow_sidx_t udp_flow_from_sock(const struct ctx *c, uint8_t pif, const union inany_addr *dst, in_port_t port, const union sockaddr_inany *s_in, const struct timespec *now); -flow_sidx_t udp_flow_from_tap(const struct ctx *c, +flow_sidx_t udp_flow_from_tap(const struct ctx *c, int queue, uint8_t pif, sa_family_t af, const void *saddr, const void *daddr, in_port_t srcport, in_port_t dstport, diff --git a/udp_vu.c b/udp_vu.c index 099677f914e7..cd2c9c516d44 100644 --- a/udp_vu.c +++ b/udp_vu.c @@ -202,9 +202,11 @@ static void udp_vu_csum(const struct flowside *toside, int iov_used) void udp_vu_sock_to_tap(const struct ctx *c, int s, int n, flow_sidx_t tosidx) { const struct flowside *toside = flowside_at_sidx(tosidx); + const struct udp_flow *uflow = udp_at_sidx(tosidx); bool v6 = !(inany_v4(&toside->eaddr) && inany_v4(&toside->oaddr)); + int rx_queue = flow_rx_virtqueue(&uflow->f); struct vu_dev *vdev = c->vdev; - struct vu_virtq *vq = &vdev->vq[VHOST_USER_RX_QUEUE]; + struct vu_virtq *vq = &vdev->vq[rx_queue]; int i;

for (i = 0; i < n; i++) { diff --git a/vu_common.c b/vu_common.c index 8904403e66af..56f26317b192 100644 --- a/vu_common.c +++ b/vu_common.c @@ -196,11 +196,11 @@ static void vu_handle_tx(struct vu_dev *vdev, int index,

data = IOV_TAIL(elem[count].out_sg, elem[count].out_num, 0); if (IOV_DROP_HEADER(&data, struct virtio_net_hdr_mrg_rxbuf)) - tap_add_packet(vdev->context, &data, now); + tap_add_packet(vdev->context, index, &data, now);

count++; } - tap_handler(vdev->context, now); + tap_handler(vdev->context, index, now);

if (count) { int i; -- 2.51.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson