#include <errno.h>
#include <unistd.h>
#include <strings.h>
+#include <malloc.h>
#include <rte_ethdev.h>
#include <rte_memcpy.h>
#include <rte_memory.h>
#include <rte_eal.h>
#include <rte_dev.h>
+#include <rte_net.h>
#include <rte_bus_vmbus.h>
#include <rte_spinlock.h>
#define HN_TXCOPY_THRESHOLD 512
#define HN_RXCOPY_THRESHOLD 256
-#define HN_RXQ_EVENT_DEFAULT 1024
+#define HN_RXQ_EVENT_DEFAULT 2048
struct hn_rxinfo {
uint32_t vlan_info;
hn_update_packet_stats(struct hn_stats *stats, const struct rte_mbuf *m)
{
uint32_t s = m->pkt_len;
- const struct ether_addr *ea;
+ const struct rte_ether_addr *ea;
if (s == 64) {
stats->size_bins[1]++;
stats->size_bins[0]++;
else if (s < 1519)
stats->size_bins[6]++;
- else if (s >= 1519)
+ else
stats->size_bins[7]++;
}
- ea = rte_pktmbuf_mtod(m, const struct ether_addr *);
+ ea = rte_pktmbuf_mtod(m, const struct rte_ether_addr *);
if (is_multicast_ether_addr(ea)) {
if (is_broadcast_ether_addr(ea))
stats->broadcast++;
return 0;
}
+void
+hn_tx_pool_uninit(struct rte_eth_dev *dev)
+{
+ struct hn_data *hv = dev->data->dev_private;
+
+ if (hv->tx_pool) {
+ rte_mempool_free(hv->tx_pool);
+ hv->tx_pool = NULL;
+ }
+}
+
static void hn_reset_txagg(struct hn_tx_queue *txq)
{
txq->agg_szleft = txq->agg_szmax;
struct hn_data *hv = dev->data->dev_private;
struct hn_tx_queue *txq;
uint32_t tx_free_thresh;
+ int err;
PMD_INIT_FUNC_TRACE();
hn_reset_txagg(txq);
- dev->data->tx_queues[queue_idx] = txq;
+ err = hn_vf_tx_queue_setup(dev, queue_idx, nb_desc,
+ socket_id, tx_conf);
+ if (err) {
+ rte_free(txq);
+ return err;
+ }
+ dev->data->tx_queues[queue_idx] = txq;
return 0;
}
m->port = rxq->port_id;
m->pkt_len = dlen;
m->data_len = dlen;
+ m->packet_type = rte_net_get_ptype(m, NULL,
+ RTE_PTYPE_L2_MASK |
+ RTE_PTYPE_L3_MASK |
+ RTE_PTYPE_L4_MASK);
if (info->vlan_info != HN_NDIS_VLAN_INFO_INVALID) {
m->vlan_tci = info->vlan_info;
m->ol_flags |= PKT_RX_VLAN_STRIPPED | PKT_RX_VLAN;
+
+ /* NDIS always strips tag, put it back if necessary */
+ if (!hv->vlan_strip && rte_vlan_insert(&m)) {
+ PMD_DRV_LOG(DEBUG, "vlan insert failed");
+ ++rxq->stats.errors;
+ rte_pktmbuf_free(m);
+ return;
+ }
}
if (info->csum_info != HN_NDIS_RXCSUM_INFO_INVALID) {
if (info->csum_info & (NDIS_RXCSUM_INFO_UDPCS_OK
| NDIS_RXCSUM_INFO_TCPCS_OK))
m->ol_flags |= PKT_RX_L4_CKSUM_GOOD;
+ else if (info->csum_info & (NDIS_RXCSUM_INFO_TCPCS_FAILED
+ | NDIS_RXCSUM_INFO_UDPCS_FAILED))
+ m->ol_flags |= PKT_RX_L4_CKSUM_BAD;
}
if (info->hash_info != HN_NDIS_HASH_INFO_INVALID) {
m->hash.rss = info->hash_value;
}
- PMD_RX_LOG(DEBUG, "port %u:%u RX id %" PRIu64 " size %u ol_flags %#" PRIx64,
+ PMD_RX_LOG(DEBUG,
+ "port %u:%u RX id %"PRIu64" size %u type %#x ol_flags %#"PRIx64,
rxq->port_id, rxq->queue_id, rxb->xactid,
- m->pkt_len, m->ol_flags);
+ m->pkt_len, m->packet_type, m->ol_flags);
++rxq->stats.packets;
rxq->stats.bytes += m->pkt_len;
hn_update_packet_stats(&rxq->stats, m);
if (unlikely(rte_ring_sp_enqueue(rxq->rx_ring, m) != 0)) {
- ++rxq->ring_full;
+ ++rxq->stats.ring_full;
rte_pktmbuf_free(m);
}
}
}
static void
-hn_rndis_receive(const struct rte_eth_dev *dev, struct hn_rx_queue *rxq,
+hn_rndis_receive(struct rte_eth_dev *dev, struct hn_rx_queue *rxq,
struct hn_rx_bufinfo *rxb, void *buf, uint32_t len)
{
const struct rndis_msghdr *hdr = buf;
break;
case RNDIS_INDICATE_STATUS_MSG:
- hn_rndis_link_status(rxq->hv, buf);
+ hn_rndis_link_status(dev, buf);
break;
case RNDIS_INITIALIZE_CMPLT:
hn_rx_buf_release(rxb);
}
+/*
+ * Called when NVS inband events are received.
+ * Send up a two part message with port_id and the NVS message
+ * to the pipe to the netvsc-vf-event control thread.
+ */
+static void hn_nvs_handle_notify(struct rte_eth_dev *dev,
+ const struct vmbus_chanpkt_hdr *pkt,
+ const void *data)
+{
+ const struct hn_nvs_hdr *hdr = data;
+
+ switch (hdr->type) {
+ case NVS_TYPE_TXTBL_NOTE:
+ /* Transmit indirection table has locking problems
+ * in DPDK and therefore not implemented
+ */
+ PMD_DRV_LOG(DEBUG, "host notify of transmit indirection table");
+ break;
+
+ case NVS_TYPE_VFASSOC_NOTE:
+ hn_nvs_handle_vfassoc(dev, pkt, data);
+ break;
+
+ default:
+ PMD_DRV_LOG(INFO,
+ "got notify, nvs type %u", hdr->type);
+ }
+}
+
struct hn_rx_queue *hn_rx_queue_alloc(struct hn_data *hv,
uint16_t queue_id,
unsigned int socket_id)
rxq = rte_zmalloc_socket("HN_RXQ", sizeof(*rxq),
RTE_CACHE_LINE_SIZE, socket_id);
- if (rxq) {
- rxq->hv = hv;
- rxq->chan = hv->channels[queue_id];
- rte_spinlock_init(&rxq->ring_lock);
- rxq->port_id = hv->port_id;
- rxq->queue_id = queue_id;
-
- rxq->event_sz = HN_RXQ_EVENT_DEFAULT;
- rxq->event_buf = rte_malloc_socket("RX_EVENTS",
- rxq->event_sz,
- RTE_CACHE_LINE_SIZE,
- socket_id);
- if (!rxq->event_buf) {
- rte_free(rxq);
- rxq = NULL;
- }
+ if (!rxq)
+ return NULL;
+
+ rxq->hv = hv;
+ rxq->chan = hv->channels[queue_id];
+ rte_spinlock_init(&rxq->ring_lock);
+ rxq->port_id = hv->port_id;
+ rxq->queue_id = queue_id;
+ rxq->event_sz = HN_RXQ_EVENT_DEFAULT;
+ rxq->event_buf = rte_malloc_socket("HN_EVENTS", HN_RXQ_EVENT_DEFAULT,
+ RTE_CACHE_LINE_SIZE, socket_id);
+ if (!rxq->event_buf) {
+ rte_free(rxq);
+ return NULL;
}
+
return rxq;
}
hn_dev_rx_queue_setup(struct rte_eth_dev *dev,
uint16_t queue_idx, uint16_t nb_desc,
unsigned int socket_id,
- const struct rte_eth_rxconf *rx_conf __rte_unused,
+ const struct rte_eth_rxconf *rx_conf,
struct rte_mempool *mp)
{
struct hn_data *hv = dev->data->dev_private;
- uint32_t qmax = hv->rxbuf_section_cnt;
char ring_name[RTE_RING_NAMESIZE];
struct hn_rx_queue *rxq;
unsigned int count;
- size_t size;
- int err = -ENOMEM;
+ int error = -ENOMEM;
PMD_INIT_FUNC_TRACE();
- if (nb_desc == 0 || nb_desc > qmax)
- nb_desc = qmax;
-
if (queue_idx == 0) {
rxq = hv->primary;
} else {
}
rxq->mb_pool = mp;
-
- count = rte_align32pow2(nb_desc);
- size = sizeof(struct rte_ring) + count * sizeof(void *);
- rxq->rx_ring = rte_malloc_socket("RX_RING", size,
- RTE_CACHE_LINE_SIZE,
- socket_id);
- if (!rxq->rx_ring)
- goto fail;
+ count = rte_mempool_avail_count(mp) / dev->data->nb_rx_queues;
+ if (nb_desc == 0 || nb_desc > count)
+ nb_desc = count;
/*
* Staging ring from receive event logic to rx_pkts.
*/
snprintf(ring_name, sizeof(ring_name),
"hn_rx_%u_%u", dev->data->port_id, queue_idx);
- err = rte_ring_init(rxq->rx_ring, ring_name,
- count, 0);
- if (err)
+ rxq->rx_ring = rte_ring_create(ring_name,
+ rte_align32pow2(nb_desc),
+ socket_id, 0);
+ if (!rxq->rx_ring)
+ goto fail;
+
+ error = hn_vf_rx_queue_setup(dev, queue_idx, nb_desc,
+ socket_id, rx_conf, mp);
+ if (error)
goto fail;
dev->data->rx_queues[queue_idx] = rxq;
rte_ring_free(rxq->rx_ring);
rte_free(rxq->event_buf);
rte_free(rxq);
- return -ENOMEM;
+ return error;
}
-void
-hn_dev_rx_queue_release(void *arg)
+static void
+hn_rx_queue_free(struct hn_rx_queue *rxq, bool keep_primary)
{
- struct hn_rx_queue *rxq = arg;
-
- PMD_INIT_FUNC_TRACE();
if (!rxq)
return;
rxq->rx_ring = NULL;
rxq->mb_pool = NULL;
- if (rxq != rxq->hv->primary) {
- rte_free(rxq->event_buf);
- rte_free(rxq);
- }
+ hn_vf_rx_queue_release(rxq->hv, rxq->queue_id);
+
+ /* Keep primary queue to allow for control operations */
+ if (keep_primary && rxq == rxq->hv->primary)
+ return;
+
+ rte_free(rxq->event_buf);
+ rte_free(rxq);
}
-static void
-hn_nvs_handle_notify(const struct vmbus_chanpkt_hdr *pkthdr,
- const void *data)
+void
+hn_dev_rx_queue_release(void *arg)
{
- const struct hn_nvs_hdr *hdr = data;
+ struct hn_rx_queue *rxq = arg;
- if (unlikely(vmbus_chanpkt_datalen(pkthdr) < sizeof(*hdr))) {
- PMD_DRV_LOG(ERR, "invalid nvs notify");
- return;
- }
+ PMD_INIT_FUNC_TRACE();
+
+ hn_rx_queue_free(rxq, true);
+}
+
+int
+hn_dev_tx_done_cleanup(void *arg, uint32_t free_cnt)
+{
+ struct hn_tx_queue *txq = arg;
- PMD_DRV_LOG(INFO,
- "got notify, nvs type %u", hdr->type);
+ return hn_process_events(txq->hv, txq->queue_id, free_cnt);
}
/*
* Process pending events on the channel.
* Called from both Rx queue poll and Tx cleanup
*/
-void hn_process_events(struct hn_data *hv, uint16_t queue_id)
+uint32_t hn_process_events(struct hn_data *hv, uint16_t queue_id,
+ uint32_t tx_limit)
{
struct rte_eth_dev *dev = &rte_eth_devices[hv->port_id];
struct hn_rx_queue *rxq;
+ uint32_t bytes_read = 0;
+ uint32_t tx_done = 0;
int ret = 0;
rxq = queue_id == 0 ? hv->primary : dev->data->rx_queues[queue_id];
/* If no pending data then nothing to do */
if (rte_vmbus_chan_rx_empty(rxq->chan))
- return;
+ return 0;
/*
* Since channel is shared between Rx and TX queue need to have a lock
* since DPDK does not force same CPU to be used for Rx/Tx.
*/
if (unlikely(!rte_spinlock_trylock(&rxq->ring_lock)))
- return;
+ return 0;
for (;;) {
const struct vmbus_chanpkt_hdr *pkt;
uint32_t len = rxq->event_sz;
const void *data;
+retry:
ret = rte_vmbus_chan_recv_raw(rxq->chan, rxq->event_buf, &len);
if (ret == -EAGAIN)
break; /* ring is empty */
- if (ret == -ENOBUFS) {
- /* expanded buffer needed */
- len = rte_align32pow2(len);
- PMD_DRV_LOG(DEBUG, "expand event buf to %u", len);
-
- rxq->event_buf = rte_realloc(rxq->event_buf,
- len, RTE_CACHE_LINE_SIZE);
- if (rxq->event_buf) {
- rxq->event_sz = len;
- continue;
- }
-
- rte_exit(EXIT_FAILURE, "can not expand event buf!\n");
+ if (unlikely(ret == -ENOBUFS)) {
+ /* event buffer not large enough to read ring */
+
+ PMD_DRV_LOG(DEBUG,
+ "event buffer expansion (need %u)", len);
+ rxq->event_sz = len + len / 4;
+ rxq->event_buf = rte_realloc(rxq->event_buf, rxq->event_sz,
+ RTE_CACHE_LINE_SIZE);
+ if (rxq->event_buf)
+ goto retry;
+ /* out of memory, no more events now */
+ rxq->event_sz = 0;
break;
}
- if (ret != 0) {
- PMD_DRV_LOG(ERR, "vmbus ring buffer error: %d", ret);
- break;
+ if (unlikely(ret <= 0)) {
+ /* This indicates a failure to communicate (or worse) */
+ rte_exit(EXIT_FAILURE,
+ "vmbus ring buffer error: %d", ret);
}
+ bytes_read += ret;
pkt = (const struct vmbus_chanpkt_hdr *)rxq->event_buf;
data = (char *)rxq->event_buf + vmbus_chanpkt_getlen(pkt->hlen);
switch (pkt->type) {
case VMBUS_CHANPKT_TYPE_COMP:
+ ++tx_done;
hn_nvs_handle_comp(dev, queue_id, pkt, data);
break;
break;
case VMBUS_CHANPKT_TYPE_INBAND:
- hn_nvs_handle_notify(pkt, data);
+ hn_nvs_handle_notify(dev, pkt, data);
break;
default:
PMD_DRV_LOG(ERR, "unknown chan pkt %u", pkt->type);
break;
}
+
+ if (tx_limit && tx_done >= tx_limit)
+ break;
+
+ if (rxq->rx_ring && rte_ring_full(rxq->rx_ring))
+ break;
}
+
+ if (bytes_read > 0)
+ rte_vmbus_chan_signal_read(rxq->chan, bytes_read);
+
rte_spinlock_unlock(&rxq->ring_lock);
- if (unlikely(ret != -EAGAIN))
- PMD_DRV_LOG(ERR, "channel receive failed: %d", ret);
+ return tx_done;
}
static void hn_append_to_chim(struct hn_tx_queue *txq,
struct hn_txdesc *txd;
if (rte_mempool_get(hv->tx_pool, (void **)&txd)) {
- ++txq->stats.nomemory;
+ ++txq->stats.ring_full;
PMD_TX_LOG(DEBUG, "tx pool exhausted!");
return NULL;
}
hn_xmit_pkts(void *ptxq, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
{
struct hn_tx_queue *txq = ptxq;
+ uint16_t queue_id = txq->queue_id;
struct hn_data *hv = txq->hv;
+ struct rte_eth_dev *vf_dev;
bool need_sig = false;
uint16_t nb_tx;
int ret;
if (unlikely(hv->closed))
return 0;
+ /* Transmit over VF if present and up */
+ vf_dev = hn_get_vf_dev(hv);
+
+ if (vf_dev && vf_dev->data->dev_started) {
+ void *sub_q = vf_dev->data->tx_queues[queue_id];
+
+ return (*vf_dev->tx_pkt_burst)(sub_q, tx_pkts, nb_pkts);
+ }
+
if (rte_mempool_avail_count(hv->tx_pool) <= txq->free_thresh)
- hn_process_events(hv, txq->queue_id);
+ hn_process_events(hv, txq->queue_id, 0);
for (nb_tx = 0; nb_tx < nb_pkts; nb_tx++) {
struct rte_mbuf *m = tx_pkts[nb_tx];
pkt = hn_try_txagg(hv, txq, pkt_size);
if (unlikely(!pkt))
- goto fail;
+ break;
- hn_encap(pkt, txq->queue_id, m);
+ hn_encap(pkt, queue_id, m);
hn_append_to_chim(txq, pkt, m);
rte_pktmbuf_free(m);
} else {
txd = hn_new_txd(hv, txq);
if (unlikely(!txd))
- goto fail;
+ break;
}
pkt = txd->rndis_pkt;
txd->data_size += m->pkt_len;
++txd->packets;
- hn_encap(pkt, txq->queue_id, m);
+ hn_encap(pkt, queue_id, m);
ret = hn_xmit_sg(txq, txd, m, &need_sig);
if (unlikely(ret != 0)) {
return nb_tx;
}
+static uint16_t
+hn_recv_vf(uint16_t vf_port, const struct hn_rx_queue *rxq,
+ struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
+{
+ uint16_t i, n;
+
+ if (unlikely(nb_pkts == 0))
+ return 0;
+
+ n = rte_eth_rx_burst(vf_port, rxq->queue_id, rx_pkts, nb_pkts);
+
+ /* relabel the received mbufs */
+ for (i = 0; i < n; i++)
+ rx_pkts[i]->port = rxq->port_id;
+
+ return n;
+}
+
uint16_t
hn_recv_pkts(void *prxq, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
{
struct hn_rx_queue *rxq = prxq;
struct hn_data *hv = rxq->hv;
+ struct rte_eth_dev *vf_dev;
+ uint16_t nb_rcv;
if (unlikely(hv->closed))
return 0;
- /* Get all outstanding receive completions */
- hn_process_events(hv, rxq->queue_id);
+ /* Receive from VF if present and up */
+ vf_dev = hn_get_vf_dev(hv);
+
+ /* Check for new completions */
+ if (likely(rte_ring_count(rxq->rx_ring) < nb_pkts))
+ hn_process_events(hv, rxq->queue_id, 0);
+
+ /* Always check the vmbus path for multicast and new flows */
+ nb_rcv = rte_ring_sc_dequeue_burst(rxq->rx_ring,
+ (void **)rx_pkts, nb_pkts, NULL);
+
+ /* If VF is available, check that as well */
+ if (vf_dev && vf_dev->data->dev_started)
+ nb_rcv += hn_recv_vf(vf_dev->data->port_id, rxq,
+ rx_pkts + nb_rcv, nb_pkts - nb_rcv);
+
+ return nb_rcv;
+}
+
+void
+hn_dev_free_queues(struct rte_eth_dev *dev)
+{
+ unsigned int i;
+
+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
+ struct hn_rx_queue *rxq = dev->data->rx_queues[i];
- /* Get mbufs off staging ring */
- return rte_ring_sc_dequeue_burst(rxq->rx_ring, (void **)rx_pkts,
- nb_pkts, NULL);
+ hn_rx_queue_free(rxq, false);
+ dev->data->rx_queues[i] = NULL;
+ }
+ dev->data->nb_rx_queues = 0;
+
+ for (i = 0; i < dev->data->nb_tx_queues; i++) {
+ hn_dev_tx_queue_release(dev->data->tx_queues[i]);
+ dev->data->tx_queues[i] = NULL;
+ }
+ dev->data->nb_tx_queues = 0;
}
git.droids-corp.org / dpdk.git / blobdiff