freebsd-nq/sys/dev/mlx5/mlx5_en/mlx5_en_tx.c
Hans Petter Selasky 0be5034007 Don't stall transmit queue on drops in mlx5en(4).
When a transmitted packet is dropped don't stall the transmit queue.

MFC after:		3 days
Approved by:		re (marius)
Sponsored by:		Mellanox Technologies
2018-09-06 12:19:36 +00:00

667 lines
17 KiB
C

/*-
* Copyright (c) 2015 Mellanox Technologies. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY AUTHOR AND CONTRIBUTORS `AS IS' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
#include "en.h"
#include <machine/atomic.h>
static inline bool
mlx5e_do_send_cqe(struct mlx5e_sq *sq)
{
sq->cev_counter++;
/* interleave the CQEs */
if (sq->cev_counter >= sq->cev_factor) {
sq->cev_counter = 0;
return (1);
}
return (0);
}
void
mlx5e_send_nop(struct mlx5e_sq *sq, u32 ds_cnt)
{
u16 pi = sq->pc & sq->wq.sz_m1;
struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi);
memset(&wqe->ctrl, 0, sizeof(wqe->ctrl));
wqe->ctrl.opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_NOP);
wqe->ctrl.qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
if (mlx5e_do_send_cqe(sq))
wqe->ctrl.fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
else
wqe->ctrl.fm_ce_se = 0;
/* Copy data for doorbell */
memcpy(sq->doorbell.d32, &wqe->ctrl, sizeof(sq->doorbell.d32));
sq->mbuf[pi].mbuf = NULL;
sq->mbuf[pi].num_bytes = 0;
sq->mbuf[pi].num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS);
sq->pc += sq->mbuf[pi].num_wqebbs;
}
#if (__FreeBSD_version >= 1100000)
static uint32_t mlx5e_hash_value;
static void
mlx5e_hash_init(void *arg)
{
mlx5e_hash_value = m_ether_tcpip_hash_init();
}
/* Make kernel call mlx5e_hash_init after the random stack finished initializing */
SYSINIT(mlx5e_hash_init, SI_SUB_RANDOM, SI_ORDER_ANY, &mlx5e_hash_init, NULL);
#endif
static struct mlx5e_sq *
mlx5e_select_queue(struct ifnet *ifp, struct mbuf *mb)
{
struct mlx5e_priv *priv = ifp->if_softc;
struct mlx5e_channel * volatile *ppch;
struct mlx5e_channel *pch;
u32 ch;
u32 tc;
ppch = priv->channel;
/* check if channels are successfully opened */
if (unlikely(ppch == NULL))
return (NULL);
/* obtain VLAN information if present */
if (mb->m_flags & M_VLANTAG) {
tc = (mb->m_pkthdr.ether_vtag >> 13);
if (tc >= priv->num_tc)
tc = priv->default_vlan_prio;
} else {
tc = priv->default_vlan_prio;
}
ch = priv->params.num_channels;
#ifdef RATELIMIT
if (mb->m_pkthdr.snd_tag != NULL) {
struct mlx5e_sq *sq;
/* check for route change */
if (mb->m_pkthdr.snd_tag->ifp != ifp)
return (NULL);
/* get pointer to sendqueue */
sq = container_of(mb->m_pkthdr.snd_tag,
struct mlx5e_rl_channel, m_snd_tag)->sq;
/* check if valid */
if (sq != NULL && sq->stopped == 0)
return (sq);
/* FALLTHROUGH */
}
#endif
/* check if flowid is set */
if (M_HASHTYPE_GET(mb) != M_HASHTYPE_NONE) {
#ifdef RSS
u32 temp;
if (rss_hash2bucket(mb->m_pkthdr.flowid,
M_HASHTYPE_GET(mb), &temp) == 0)
ch = temp % ch;
else
#endif
ch = (mb->m_pkthdr.flowid % 128) % ch;
} else {
#if (__FreeBSD_version >= 1100000)
ch = m_ether_tcpip_hash(MBUF_HASHFLAG_L3 |
MBUF_HASHFLAG_L4, mb, mlx5e_hash_value) % ch;
#else
/*
* m_ether_tcpip_hash not present in stable, so just
* throw unhashed mbufs on queue 0
*/
ch = 0;
#endif
}
/* check if channel is allocated and not stopped */
pch = ppch[ch];
if (likely(pch != NULL && pch->sq[tc].stopped == 0))
return (&pch->sq[tc]);
return (NULL);
}
static inline u16
mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, struct mbuf *mb)
{
switch(sq->min_inline_mode) {
case MLX5_INLINE_MODE_NONE:
/*
* When inline mode is NONE, we do not need to copy
* headers into WQEs, except when vlan tag framing is
* requested. Hardware might offload vlan tagging on
* transmit. This is a separate capability, which is
* known to be disabled on ConnectX-5 due to a hardware
* bug RM 931383. If vlan_inline_cap is not present and
* the packet has vlan tag, fall back to inlining.
*/
if ((mb->m_flags & M_VLANTAG) != 0 &&
sq->vlan_inline_cap == 0)
break;
return (0);
case MLX5_INLINE_MODE_L2:
/*
* Due to hardware limitations, when trust mode is
* DSCP, the hardware may request MLX5_INLINE_MODE_L2
* while it really needs all L2 headers and the 4 first
* bytes of the IP header (which include the
* TOS/traffic-class).
*
* To avoid doing a firmware command for querying the
* trust state and parsing the mbuf for doing
* unnecessary checks (VLAN/eth_type) in the fast path,
* we are going for the worth case (22 Bytes) if
* the mb->m_pkthdr.len allows it.
*/
if (mb->m_pkthdr.len > ETHER_HDR_LEN +
ETHER_VLAN_ENCAP_LEN + 4)
return (MIN(sq->max_inline, ETHER_HDR_LEN +
ETHER_VLAN_ENCAP_LEN + 4));
break;
}
return (MIN(sq->max_inline, mb->m_pkthdr.len));
}
static int
mlx5e_get_header_size(struct mbuf *mb)
{
struct ether_vlan_header *eh;
struct tcphdr *th;
struct ip *ip;
int ip_hlen, tcp_hlen;
struct ip6_hdr *ip6;
uint16_t eth_type;
int eth_hdr_len;
eh = mtod(mb, struct ether_vlan_header *);
if (mb->m_len < ETHER_HDR_LEN)
return (0);
if (eh->evl_encap_proto == htons(ETHERTYPE_VLAN)) {
eth_type = ntohs(eh->evl_proto);
eth_hdr_len = ETHER_HDR_LEN + ETHER_VLAN_ENCAP_LEN;
} else {
eth_type = ntohs(eh->evl_encap_proto);
eth_hdr_len = ETHER_HDR_LEN;
}
if (mb->m_len < eth_hdr_len)
return (0);
switch (eth_type) {
case ETHERTYPE_IP:
ip = (struct ip *)(mb->m_data + eth_hdr_len);
if (mb->m_len < eth_hdr_len + sizeof(*ip))
return (0);
if (ip->ip_p != IPPROTO_TCP)
return (0);
ip_hlen = ip->ip_hl << 2;
eth_hdr_len += ip_hlen;
break;
case ETHERTYPE_IPV6:
ip6 = (struct ip6_hdr *)(mb->m_data + eth_hdr_len);
if (mb->m_len < eth_hdr_len + sizeof(*ip6))
return (0);
if (ip6->ip6_nxt != IPPROTO_TCP)
return (0);
eth_hdr_len += sizeof(*ip6);
break;
default:
return (0);
}
if (mb->m_len < eth_hdr_len + sizeof(*th))
return (0);
th = (struct tcphdr *)(mb->m_data + eth_hdr_len);
tcp_hlen = th->th_off << 2;
eth_hdr_len += tcp_hlen;
if (mb->m_len < eth_hdr_len)
return (0);
return (eth_hdr_len);
}
/*
* The return value is not going back to the stack because of
* the drbr
*/
static int
mlx5e_sq_xmit(struct mlx5e_sq *sq, struct mbuf **mbp)
{
bus_dma_segment_t segs[MLX5E_MAX_TX_MBUF_FRAGS];
struct mlx5_wqe_data_seg *dseg;
struct mlx5e_tx_wqe *wqe;
struct ifnet *ifp;
int nsegs;
int err;
int x;
struct mbuf *mb = *mbp;
u16 ds_cnt;
u16 ihs;
u16 pi;
u8 opcode;
/*
* Return ENOBUFS if the queue is full, this may trigger reinsertion
* of the mbuf into the drbr (see mlx5e_xmit_locked)
*/
if (unlikely(!mlx5e_sq_has_room_for(sq, 2 * MLX5_SEND_WQE_MAX_WQEBBS))) {
return (ENOBUFS);
}
/* Align SQ edge with NOPs to avoid WQE wrap around */
pi = ((~sq->pc) & sq->wq.sz_m1);
if (pi < (MLX5_SEND_WQE_MAX_WQEBBS - 1)) {
/* Send one multi NOP message instead of many */
mlx5e_send_nop(sq, (pi + 1) * MLX5_SEND_WQEBB_NUM_DS);
pi = ((~sq->pc) & sq->wq.sz_m1);
if (pi < (MLX5_SEND_WQE_MAX_WQEBBS - 1))
return (ENOMEM);
}
/* Setup local variables */
pi = sq->pc & sq->wq.sz_m1;
wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi);
ifp = sq->ifp;
memset(wqe, 0, sizeof(*wqe));
/* Send a copy of the frame to the BPF listener, if any */
if (ifp != NULL && ifp->if_bpf != NULL)
ETHER_BPF_MTAP(ifp, mb);
if (mb->m_pkthdr.csum_flags & (CSUM_IP | CSUM_TSO)) {
wqe->eth.cs_flags |= MLX5_ETH_WQE_L3_CSUM;
}
if (mb->m_pkthdr.csum_flags & (CSUM_TCP | CSUM_UDP | CSUM_UDP_IPV6 | CSUM_TCP_IPV6 | CSUM_TSO)) {
wqe->eth.cs_flags |= MLX5_ETH_WQE_L4_CSUM;
}
if (wqe->eth.cs_flags == 0) {
sq->stats.csum_offload_none++;
}
if (mb->m_pkthdr.csum_flags & CSUM_TSO) {
u32 payload_len;
u32 mss = mb->m_pkthdr.tso_segsz;
u32 num_pkts;
wqe->eth.mss = cpu_to_be16(mss);
opcode = MLX5_OPCODE_LSO;
ihs = mlx5e_get_header_size(mb);
payload_len = mb->m_pkthdr.len - ihs;
if (payload_len == 0)
num_pkts = 1;
else
num_pkts = DIV_ROUND_UP(payload_len, mss);
sq->mbuf[pi].num_bytes = payload_len + (num_pkts * ihs);
sq->stats.tso_packets++;
sq->stats.tso_bytes += payload_len;
} else {
opcode = MLX5_OPCODE_SEND;
ihs = mlx5e_get_inline_hdr_size(sq, mb);
sq->mbuf[pi].num_bytes = max_t (unsigned int,
mb->m_pkthdr.len, ETHER_MIN_LEN - ETHER_CRC_LEN);
}
if (ihs == 0) {
if ((mb->m_flags & M_VLANTAG) != 0) {
wqe->eth.vlan_cmd = htons(0x8000); /* bit 0 CVLAN */
wqe->eth.vlan_hdr = htons(mb->m_pkthdr.ether_vtag);
} else {
wqe->eth.inline_hdr_sz = 0;
}
} else {
if ((mb->m_flags & M_VLANTAG) != 0) {
struct ether_vlan_header *eh = (struct ether_vlan_header
*)wqe->eth.inline_hdr_start;
/* Range checks */
if (ihs > (MLX5E_MAX_TX_INLINE - ETHER_VLAN_ENCAP_LEN))
ihs = (MLX5E_MAX_TX_INLINE -
ETHER_VLAN_ENCAP_LEN);
else if (ihs < ETHER_HDR_LEN) {
err = EINVAL;
goto tx_drop;
}
m_copydata(mb, 0, ETHER_HDR_LEN, (caddr_t)eh);
m_adj(mb, ETHER_HDR_LEN);
/* Insert 4 bytes VLAN tag into data stream */
eh->evl_proto = eh->evl_encap_proto;
eh->evl_encap_proto = htons(ETHERTYPE_VLAN);
eh->evl_tag = htons(mb->m_pkthdr.ether_vtag);
/* Copy rest of header data, if any */
m_copydata(mb, 0, ihs - ETHER_HDR_LEN, (caddr_t)(eh +
1));
m_adj(mb, ihs - ETHER_HDR_LEN);
/* Extend header by 4 bytes */
ihs += ETHER_VLAN_ENCAP_LEN;
} else {
m_copydata(mb, 0, ihs, wqe->eth.inline_hdr_start);
m_adj(mb, ihs);
}
wqe->eth.inline_hdr_sz = cpu_to_be16(ihs);
}
ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS;
if (ihs > sizeof(wqe->eth.inline_hdr_start)) {
ds_cnt += DIV_ROUND_UP(ihs - sizeof(wqe->eth.inline_hdr_start),
MLX5_SEND_WQE_DS);
}
dseg = ((struct mlx5_wqe_data_seg *)&wqe->ctrl) + ds_cnt;
err = bus_dmamap_load_mbuf_sg(sq->dma_tag, sq->mbuf[pi].dma_map,
mb, segs, &nsegs, BUS_DMA_NOWAIT);
if (err == EFBIG) {
/* Update statistics */
sq->stats.defragged++;
/* Too many mbuf fragments */
mb = m_defrag(*mbp, M_NOWAIT);
if (mb == NULL) {
mb = *mbp;
goto tx_drop;
}
/* Try again */
err = bus_dmamap_load_mbuf_sg(sq->dma_tag, sq->mbuf[pi].dma_map,
mb, segs, &nsegs, BUS_DMA_NOWAIT);
}
/* Catch errors */
if (err != 0)
goto tx_drop;
/* Make sure all mbuf data, if any, is written to RAM */
if (nsegs != 0) {
bus_dmamap_sync(sq->dma_tag, sq->mbuf[pi].dma_map,
BUS_DMASYNC_PREWRITE);
} else {
/* All data was inlined, free the mbuf. */
bus_dmamap_unload(sq->dma_tag, sq->mbuf[pi].dma_map);
m_freem(mb);
mb = NULL;
}
for (x = 0; x != nsegs; x++) {
if (segs[x].ds_len == 0)
continue;
dseg->addr = cpu_to_be64((uint64_t)segs[x].ds_addr);
dseg->lkey = sq->mkey_be;
dseg->byte_count = cpu_to_be32((uint32_t)segs[x].ds_len);
dseg++;
}
ds_cnt = (dseg - ((struct mlx5_wqe_data_seg *)&wqe->ctrl));
wqe->ctrl.opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | opcode);
wqe->ctrl.qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt);
if (mlx5e_do_send_cqe(sq))
wqe->ctrl.fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
else
wqe->ctrl.fm_ce_se = 0;
/* Copy data for doorbell */
memcpy(sq->doorbell.d32, &wqe->ctrl, sizeof(sq->doorbell.d32));
/* Store pointer to mbuf */
sq->mbuf[pi].mbuf = mb;
sq->mbuf[pi].num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS);
sq->pc += sq->mbuf[pi].num_wqebbs;
sq->stats.packets++;
*mbp = NULL; /* safety clear */
return (0);
tx_drop:
sq->stats.dropped++;
*mbp = NULL;
m_freem(mb);
return err;
}
static void
mlx5e_poll_tx_cq(struct mlx5e_sq *sq, int budget)
{
u16 sqcc;
/*
* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
* otherwise a cq overrun may occur
*/
sqcc = sq->cc;
while (budget > 0) {
struct mlx5_cqe64 *cqe;
struct mbuf *mb;
u16 x;
u16 ci;
cqe = mlx5e_get_cqe(&sq->cq);
if (!cqe)
break;
mlx5_cqwq_pop(&sq->cq.wq);
/* update budget according to the event factor */
budget -= sq->cev_factor;
for (x = 0; x != sq->cev_factor; x++) {
ci = sqcc & sq->wq.sz_m1;
mb = sq->mbuf[ci].mbuf;
sq->mbuf[ci].mbuf = NULL; /* Safety clear */
if (mb == NULL) {
if (sq->mbuf[ci].num_bytes == 0) {
/* NOP */
sq->stats.nop++;
}
} else {
bus_dmamap_sync(sq->dma_tag, sq->mbuf[ci].dma_map,
BUS_DMASYNC_POSTWRITE);
bus_dmamap_unload(sq->dma_tag, sq->mbuf[ci].dma_map);
/* Free transmitted mbuf */
m_freem(mb);
}
sqcc += sq->mbuf[ci].num_wqebbs;
}
}
mlx5_cqwq_update_db_record(&sq->cq.wq);
/* Ensure cq space is freed before enabling more cqes */
atomic_thread_fence_rel();
sq->cc = sqcc;
if (sq->sq_tq != NULL &&
atomic_cmpset_int(&sq->queue_state, MLX5E_SQ_FULL, MLX5E_SQ_READY))
taskqueue_enqueue(sq->sq_tq, &sq->sq_task);
}
static int
mlx5e_xmit_locked(struct ifnet *ifp, struct mlx5e_sq *sq, struct mbuf *mb)
{
struct mbuf *next;
int err = 0;
if (likely(mb != NULL)) {
/*
* If we can't insert mbuf into drbr, try to xmit anyway.
* We keep the error we got so we could return that after xmit.
*/
err = drbr_enqueue(ifp, sq->br, mb);
}
/*
* Check if the network interface is closed or if the SQ is
* being stopped:
*/
if (unlikely((ifp->if_drv_flags & IFF_DRV_RUNNING) == 0 ||
sq->stopped != 0))
return (err);
/* Process the queue */
while ((next = drbr_peek(ifp, sq->br)) != NULL) {
if (mlx5e_sq_xmit(sq, &next) != 0) {
if (next != NULL) {
drbr_putback(ifp, sq->br, next);
atomic_store_rel_int(&sq->queue_state, MLX5E_SQ_FULL);
break;
}
}
drbr_advance(ifp, sq->br);
}
/* Check if we need to write the doorbell */
if (likely(sq->doorbell.d64 != 0)) {
mlx5e_tx_notify_hw(sq, sq->doorbell.d32, 0);
sq->doorbell.d64 = 0;
}
/*
* Check if we need to start the event timer which flushes the
* transmit ring on timeout:
*/
if (unlikely(sq->cev_next_state == MLX5E_CEV_STATE_INITIAL &&
sq->cev_factor != 1)) {
/* start the timer */
mlx5e_sq_cev_timeout(sq);
} else {
/* don't send NOPs yet */
sq->cev_next_state = MLX5E_CEV_STATE_HOLD_NOPS;
}
return (err);
}
static int
mlx5e_xmit_locked_no_br(struct ifnet *ifp, struct mlx5e_sq *sq, struct mbuf *mb)
{
int err = 0;
if (unlikely((ifp->if_drv_flags & IFF_DRV_RUNNING) == 0 ||
sq->stopped != 0)) {
m_freem(mb);
return (ENETDOWN);
}
/* Do transmit */
if (mlx5e_sq_xmit(sq, &mb) != 0) {
/* NOTE: m_freem() is NULL safe */
m_freem(mb);
err = ENOBUFS;
}
/* Check if we need to write the doorbell */
if (likely(sq->doorbell.d64 != 0)) {
mlx5e_tx_notify_hw(sq, sq->doorbell.d32, 0);
sq->doorbell.d64 = 0;
}
/*
* Check if we need to start the event timer which flushes the
* transmit ring on timeout:
*/
if (unlikely(sq->cev_next_state == MLX5E_CEV_STATE_INITIAL &&
sq->cev_factor != 1)) {
/* start the timer */
mlx5e_sq_cev_timeout(sq);
} else {
/* don't send NOPs yet */
sq->cev_next_state = MLX5E_CEV_STATE_HOLD_NOPS;
}
return (err);
}
int
mlx5e_xmit(struct ifnet *ifp, struct mbuf *mb)
{
struct mlx5e_sq *sq;
int ret;
sq = mlx5e_select_queue(ifp, mb);
if (unlikely(sq == NULL)) {
#ifdef RATELIMIT
/* Check for route change */
if (mb->m_pkthdr.snd_tag != NULL &&
mb->m_pkthdr.snd_tag->ifp != ifp) {
/* Free mbuf */
m_freem(mb);
/*
* Tell upper layers about route change and to
* re-transmit this packet:
*/
return (EAGAIN);
}
#endif
/* Free mbuf */
m_freem(mb);
/* Invalid send queue */
return (ENXIO);
}
if (unlikely(sq->br == NULL)) {
/* rate limited traffic */
mtx_lock(&sq->lock);
ret = mlx5e_xmit_locked_no_br(ifp, sq, mb);
mtx_unlock(&sq->lock);
} else if (mtx_trylock(&sq->lock)) {
ret = mlx5e_xmit_locked(ifp, sq, mb);
mtx_unlock(&sq->lock);
} else {
ret = drbr_enqueue(ifp, sq->br, mb);
taskqueue_enqueue(sq->sq_tq, &sq->sq_task);
}
return (ret);
}
void
mlx5e_tx_cq_comp(struct mlx5_core_cq *mcq)
{
struct mlx5e_sq *sq = container_of(mcq, struct mlx5e_sq, cq.mcq);
mtx_lock(&sq->comp_lock);
mlx5e_poll_tx_cq(sq, MLX5E_BUDGET_MAX);
mlx5e_cq_arm(&sq->cq, MLX5_GET_DOORBELL_LOCK(&sq->priv->doorbell_lock));
mtx_unlock(&sq->comp_lock);
}
void
mlx5e_tx_que(void *context, int pending)
{
struct mlx5e_sq *sq = context;
struct ifnet *ifp = sq->ifp;
if (ifp->if_drv_flags & IFF_DRV_RUNNING) {
mtx_lock(&sq->lock);
if (!drbr_empty(ifp, sq->br))
mlx5e_xmit_locked(ifp, sq, NULL);
mtx_unlock(&sq->lock);
}
}