X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Fdpaa%2Fdpaa_rxtx.c;h=30b1836079cce3c237bc09924e450e8065e79d72;hb=62d6f70f30f4eaeb533b37b539ffebd2398e0cfe;hp=3e3719d1dd53513e779ff9ae8eefe19011d1938b;hpb=43797e7b47741d58e5061255171f82a9d423bdf4;p=dpdk.git

diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/drivers/net/dpaa/dpaa_rxtx.c
index 3e3719d1dd..30b1836079 100644
--- a/drivers/net/dpaa/dpaa_rxtx.c
+++ b/drivers/net/dpaa/dpaa_rxtx.c
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause
  *
  *   Copyright 2016 Freescale Semiconductor, Inc. All rights reserved.
- *   Copyright 2017 NXP
+ *   Copyright 2017,2019 NXP
  *
  */
 
@@ -26,19 +26,21 @@
 #include <rte_eal.h>
 #include <rte_alarm.h>
 #include <rte_ether.h>
-#include <rte_ethdev.h>
+#include <rte_ethdev_driver.h>
 #include <rte_malloc.h>
 #include <rte_ring.h>
 #include <rte_ip.h>
 #include <rte_tcp.h>
 #include <rte_udp.h>
 #include <rte_net.h>
+#include <rte_eventdev.h>
 
 #include "dpaa_ethdev.h"
 #include "dpaa_rxtx.h"
 #include <rte_dpaa_bus.h>
 #include <dpaa_mempool.h>
 
+#include <qman.h>
 #include <fsl_usd.h>
 #include <fsl_qman.h>
 #include <fsl_bman.h>
@@ -57,7 +59,7 @@
 	} while (0)
 
 #if (defined RTE_LIBRTE_DPAA_DEBUG_DRIVER)
-void dpaa_display_frame(const struct qm_fd *fd)
+static void dpaa_display_frame(const struct qm_fd *fd)
 {
 	int ii;
 	char *ptr;
@@ -88,11 +90,10 @@ static inline void dpaa_slow_parsing(struct rte_mbuf *m __rte_unused,
 	/*TBD:XXX: to be implemented*/
 }
 
-static inline void dpaa_eth_packet_info(struct rte_mbuf *m,
-					uint64_t fd_virt_addr)
+static inline void dpaa_eth_packet_info(struct rte_mbuf *m, void *fd_virt_addr)
 {
 	struct annotations_t *annot = GET_ANNOTATIONS(fd_virt_addr);
-	uint64_t prs = *((uint64_t *)(&annot->parse)) & DPAA_PARSE_MASK;
+	uint64_t prs = *((uintptr_t *)(&annot->parse)) & DPAA_PARSE_MASK;
 
 	DPAA_DP_LOG(DEBUG, " Parsing mbuf: %p with annotations: %p", m, annot);
 
@@ -197,44 +198,45 @@ static inline void dpaa_eth_packet_info(struct rte_mbuf *m,
 
 static inline void dpaa_checksum(struct rte_mbuf *mbuf)
 {
-	struct ether_hdr *eth_hdr = rte_pktmbuf_mtod(mbuf, struct ether_hdr *);
+	struct rte_ether_hdr *eth_hdr =
+		rte_pktmbuf_mtod(mbuf, struct rte_ether_hdr *);
 	char *l3_hdr = (char *)eth_hdr + mbuf->l2_len;
-	struct ipv4_hdr *ipv4_hdr = (struct ipv4_hdr *)l3_hdr;
-	struct ipv6_hdr *ipv6_hdr = (struct ipv6_hdr *)l3_hdr;
+	struct rte_ipv4_hdr *ipv4_hdr = (struct rte_ipv4_hdr *)l3_hdr;
+	struct rte_ipv6_hdr *ipv6_hdr = (struct rte_ipv6_hdr *)l3_hdr;
 
 	DPAA_DP_LOG(DEBUG, "Calculating checksum for mbuf: %p", mbuf);
 
 	if (((mbuf->packet_type & RTE_PTYPE_L3_MASK) == RTE_PTYPE_L3_IPV4) ||
 	    ((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
 	    RTE_PTYPE_L3_IPV4_EXT)) {
-		ipv4_hdr = (struct ipv4_hdr *)l3_hdr;
+		ipv4_hdr = (struct rte_ipv4_hdr *)l3_hdr;
 		ipv4_hdr->hdr_checksum = 0;
 		ipv4_hdr->hdr_checksum = rte_ipv4_cksum(ipv4_hdr);
 	} else if (((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
 		   RTE_PTYPE_L3_IPV6) ||
 		   ((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
 		   RTE_PTYPE_L3_IPV6_EXT))
-		ipv6_hdr = (struct ipv6_hdr *)l3_hdr;
+		ipv6_hdr = (struct rte_ipv6_hdr *)l3_hdr;
 
 	if ((mbuf->packet_type & RTE_PTYPE_L4_MASK) == RTE_PTYPE_L4_TCP) {
-		struct tcp_hdr *tcp_hdr = (struct tcp_hdr *)(l3_hdr +
+		struct rte_tcp_hdr *tcp_hdr = (struct rte_tcp_hdr *)(l3_hdr +
 					  mbuf->l3_len);
 		tcp_hdr->cksum = 0;
-		if (eth_hdr->ether_type == htons(ETHER_TYPE_IPv4))
+		if (eth_hdr->ether_type == htons(RTE_ETHER_TYPE_IPV4))
 			tcp_hdr->cksum = rte_ipv4_udptcp_cksum(ipv4_hdr,
 							       tcp_hdr);
-		else /* assume ethertype == ETHER_TYPE_IPv6 */
+		else /* assume ethertype == RTE_ETHER_TYPE_IPV6 */
 			tcp_hdr->cksum = rte_ipv6_udptcp_cksum(ipv6_hdr,
 							       tcp_hdr);
 	} else if ((mbuf->packet_type & RTE_PTYPE_L4_MASK) ==
 		   RTE_PTYPE_L4_UDP) {
-		struct udp_hdr *udp_hdr = (struct udp_hdr *)(l3_hdr +
+		struct rte_udp_hdr *udp_hdr = (struct rte_udp_hdr *)(l3_hdr +
 							     mbuf->l3_len);
 		udp_hdr->dgram_cksum = 0;
-		if (eth_hdr->ether_type == htons(ETHER_TYPE_IPv4))
+		if (eth_hdr->ether_type == htons(RTE_ETHER_TYPE_IPV4))
 			udp_hdr->dgram_cksum = rte_ipv4_udptcp_cksum(ipv4_hdr,
 								     udp_hdr);
-		else /* assume ethertype == ETHER_TYPE_IPv6 */
+		else /* assume ethertype == RTE_ETHER_TYPE_IPV6 */
 			udp_hdr->dgram_cksum = rte_ipv6_udptcp_cksum(ipv6_hdr,
 								     udp_hdr);
 	}
@@ -305,9 +307,7 @@ dpaa_eth_sg_to_mbuf(const struct qm_fd *fd, uint32_t ifid)
 	int i = 0;
 	uint8_t fd_offset = fd->offset;
 
-	DPAA_DP_LOG(DEBUG, "Received an SG frame");
-
-	vaddr = rte_dpaa_mem_ptov(qm_fd_addr(fd));
+	vaddr = DPAA_MEMPOOL_PTOV(bp_info, qm_fd_addr(fd));
 	if (!vaddr) {
 		DPAA_PMD_ERR("unable to convert physical address");
 		return NULL;
@@ -316,7 +316,7 @@ dpaa_eth_sg_to_mbuf(const struct qm_fd *fd, uint32_t ifid)
 	sg_temp = &sgt[i++];
 	hw_sg_to_cpu(sg_temp);
 	temp = (struct rte_mbuf *)((char *)vaddr - bp_info->meta_data_size);
-	sg_vaddr = rte_dpaa_mem_ptov(qm_sg_entry_get64(sg_temp));
+	sg_vaddr = DPAA_MEMPOOL_PTOV(bp_info, qm_sg_entry_get64(sg_temp));
 
 	first_seg = (struct rte_mbuf *)((char *)sg_vaddr -
 						bp_info->meta_data_size);
@@ -332,7 +332,8 @@ dpaa_eth_sg_to_mbuf(const struct qm_fd *fd, uint32_t ifid)
 	while (i < DPAA_SGT_MAX_ENTRIES) {
 		sg_temp = &sgt[i++];
 		hw_sg_to_cpu(sg_temp);
-		sg_vaddr = rte_dpaa_mem_ptov(qm_sg_entry_get64(sg_temp));
+		sg_vaddr = DPAA_MEMPOOL_PTOV(bp_info,
+					     qm_sg_entry_get64(sg_temp));
 		cur_seg = (struct rte_mbuf *)((char *)sg_vaddr -
 						      bp_info->meta_data_size);
 		cur_seg->data_off = sg_temp->offset;
@@ -347,8 +348,10 @@ dpaa_eth_sg_to_mbuf(const struct qm_fd *fd, uint32_t ifid)
 		}
 		prev_seg = cur_seg;
 	}
+	DPAA_DP_LOG(DEBUG, "Received an SG frame len =%d, num_sg =%d",
+			first_seg->pkt_len, first_seg->nb_segs);
 
-	dpaa_eth_packet_info(first_seg, (uint64_t)vaddr);
+	dpaa_eth_packet_info(first_seg, vaddr);
 	rte_pktmbuf_free_seg(temp);
 
 	return first_seg;
@@ -359,26 +362,27 @@ dpaa_eth_fd_to_mbuf(const struct qm_fd *fd, uint32_t ifid)
 {
 	struct rte_mbuf *mbuf;
 	struct dpaa_bp_info *bp_info = DPAA_BPID_TO_POOL_INFO(fd->bpid);
-	void *ptr = rte_dpaa_mem_ptov(qm_fd_addr(fd));
+	void *ptr;
 	uint8_t format =
 		(fd->opaque & DPAA_FD_FORMAT_MASK) >> DPAA_FD_FORMAT_SHIFT;
 	uint16_t offset;
 	uint32_t length;
 
-	DPAA_DP_LOG(DEBUG, " FD--->MBUF");
-
 	if (unlikely(format == qm_fd_sg))
 		return dpaa_eth_sg_to_mbuf(fd, ifid);
 
-	rte_prefetch0((void *)((uint8_t *)ptr + DEFAULT_RX_ICEOF));
-
 	offset = (fd->opaque & DPAA_FD_OFFSET_MASK) >> DPAA_FD_OFFSET_SHIFT;
 	length = fd->opaque & DPAA_FD_LENGTH_MASK;
 
+	DPAA_DP_LOG(DEBUG, " FD--->MBUF off %d len = %d", offset, length);
+
 	/* Ignoring case when format != qm_fd_contig */
 	dpaa_display_frame(fd);
+	ptr = DPAA_MEMPOOL_PTOV(bp_info, qm_fd_addr(fd));
 
 	mbuf = (struct rte_mbuf *)((char *)ptr - bp_info->meta_data_size);
+	/* Prefetch the Parse results and packet data to L1 */
+	rte_prefetch0((void *)((uint8_t *)ptr + DEFAULT_RX_ICEOF));
 
 	mbuf->data_off = offset;
 	mbuf->data_len = length;
@@ -389,22 +393,121 @@ dpaa_eth_fd_to_mbuf(const struct qm_fd *fd, uint32_t ifid)
 	mbuf->ol_flags = 0;
 	mbuf->next = NULL;
 	rte_mbuf_refcnt_set(mbuf, 1);
-	dpaa_eth_packet_info(mbuf, (uint64_t)mbuf->buf_addr);
+	dpaa_eth_packet_info(mbuf, mbuf->buf_addr);
 
 	return mbuf;
 }
 
-enum qman_cb_dqrr_result dpaa_rx_cb(void *event __always_unused,
-				    struct qman_portal *qm __always_unused,
-				    struct qman_fq *fq,
-				    const struct qm_dqrr_entry *dqrr,
-				    void **bufs)
+/* Specific for LS1043 */
+void
+dpaa_rx_cb_no_prefetch(struct qman_fq **fq, struct qm_dqrr_entry **dqrr,
+	   void **bufs, int num_bufs)
 {
-	const struct qm_fd *fd = &dqrr->fd;
+	struct rte_mbuf *mbuf;
+	struct dpaa_bp_info *bp_info;
+	const struct qm_fd *fd;
+	void *ptr;
+	struct dpaa_if *dpaa_intf;
+	uint16_t offset, i;
+	uint32_t length;
+	uint8_t format;
 
-	*bufs = dpaa_eth_fd_to_mbuf(fd,
-			((struct dpaa_if *)fq->dpaa_intf)->ifid);
-	return qman_cb_dqrr_consume;
+	bp_info = DPAA_BPID_TO_POOL_INFO(dqrr[0]->fd.bpid);
+	ptr = rte_dpaa_mem_ptov(qm_fd_addr(&dqrr[0]->fd));
+	rte_prefetch0((void *)((uint8_t *)ptr + DEFAULT_RX_ICEOF));
+	bufs[0] = (struct rte_mbuf *)((char *)ptr - bp_info->meta_data_size);
+
+	for (i = 0; i < num_bufs; i++) {
+		if (i < num_bufs - 1) {
+			bp_info = DPAA_BPID_TO_POOL_INFO(dqrr[i + 1]->fd.bpid);
+			ptr = rte_dpaa_mem_ptov(qm_fd_addr(&dqrr[i + 1]->fd));
+			rte_prefetch0((void *)((uint8_t *)ptr +
+					DEFAULT_RX_ICEOF));
+			bufs[i + 1] = (struct rte_mbuf *)((char *)ptr -
+					bp_info->meta_data_size);
+		}
+
+		fd = &dqrr[i]->fd;
+		dpaa_intf = fq[0]->dpaa_intf;
+
+		format = (fd->opaque & DPAA_FD_FORMAT_MASK) >>
+				DPAA_FD_FORMAT_SHIFT;
+		if (unlikely(format == qm_fd_sg)) {
+			bufs[i] = dpaa_eth_sg_to_mbuf(fd, dpaa_intf->ifid);
+			continue;
+		}
+
+		offset = (fd->opaque & DPAA_FD_OFFSET_MASK) >>
+				DPAA_FD_OFFSET_SHIFT;
+		length = fd->opaque & DPAA_FD_LENGTH_MASK;
+
+		mbuf = bufs[i];
+		mbuf->data_off = offset;
+		mbuf->data_len = length;
+		mbuf->pkt_len = length;
+		mbuf->port = dpaa_intf->ifid;
+
+		mbuf->nb_segs = 1;
+		mbuf->ol_flags = 0;
+		mbuf->next = NULL;
+		rte_mbuf_refcnt_set(mbuf, 1);
+		dpaa_eth_packet_info(mbuf, mbuf->buf_addr);
+	}
+}
+
+void
+dpaa_rx_cb(struct qman_fq **fq, struct qm_dqrr_entry **dqrr,
+	   void **bufs, int num_bufs)
+{
+	struct rte_mbuf *mbuf;
+	const struct qm_fd *fd;
+	struct dpaa_if *dpaa_intf;
+	uint16_t offset, i;
+	uint32_t length;
+	uint8_t format;
+
+	for (i = 0; i < num_bufs; i++) {
+		fd = &dqrr[i]->fd;
+		dpaa_intf = fq[0]->dpaa_intf;
+
+		format = (fd->opaque & DPAA_FD_FORMAT_MASK) >>
+				DPAA_FD_FORMAT_SHIFT;
+		if (unlikely(format == qm_fd_sg)) {
+			bufs[i] = dpaa_eth_sg_to_mbuf(fd, dpaa_intf->ifid);
+			continue;
+		}
+
+		offset = (fd->opaque & DPAA_FD_OFFSET_MASK) >>
+				DPAA_FD_OFFSET_SHIFT;
+		length = fd->opaque & DPAA_FD_LENGTH_MASK;
+
+		mbuf = bufs[i];
+		mbuf->data_off = offset;
+		mbuf->data_len = length;
+		mbuf->pkt_len = length;
+		mbuf->port = dpaa_intf->ifid;
+
+		mbuf->nb_segs = 1;
+		mbuf->ol_flags = 0;
+		mbuf->next = NULL;
+		rte_mbuf_refcnt_set(mbuf, 1);
+		dpaa_eth_packet_info(mbuf, mbuf->buf_addr);
+	}
+}
+
+void dpaa_rx_cb_prepare(struct qm_dqrr_entry *dq, void **bufs)
+{
+	struct dpaa_bp_info *bp_info = DPAA_BPID_TO_POOL_INFO(dq->fd.bpid);
+	void *ptr = rte_dpaa_mem_ptov(qm_fd_addr(&dq->fd));
+
+	/* In case of LS1046, annotation stashing is disabled due to L2 cache
+	 * being bottleneck in case of multicore scanario for this platform.
+	 * So we prefetch the annoation beforehand, so that it is available
+	 * in cache when accessed.
+	 */
+	rte_prefetch0((void *)((uint8_t *)ptr + DEFAULT_RX_ICEOF));
+
+	*bufs = (struct rte_mbuf *)((char *)ptr - bp_info->meta_data_size);
 }
 
 static uint16_t
@@ -425,6 +528,67 @@ dpaa_eth_queue_portal_rx(struct qman_fq *fq,
 	return qman_portal_poll_rx(nb_bufs, (void **)bufs, fq->qp);
 }
 
+enum qman_cb_dqrr_result
+dpaa_rx_cb_parallel(void *event,
+		    struct qman_portal *qm __always_unused,
+		    struct qman_fq *fq,
+		    const struct qm_dqrr_entry *dqrr,
+		    void **bufs)
+{
+	u32 ifid = ((struct dpaa_if *)fq->dpaa_intf)->ifid;
+	struct rte_mbuf *mbuf;
+	struct rte_event *ev = (struct rte_event *)event;
+
+	mbuf = dpaa_eth_fd_to_mbuf(&dqrr->fd, ifid);
+	ev->event_ptr = (void *)mbuf;
+	ev->flow_id = fq->ev.flow_id;
+	ev->sub_event_type = fq->ev.sub_event_type;
+	ev->event_type = RTE_EVENT_TYPE_ETHDEV;
+	ev->op = RTE_EVENT_OP_NEW;
+	ev->sched_type = fq->ev.sched_type;
+	ev->queue_id = fq->ev.queue_id;
+	ev->priority = fq->ev.priority;
+	ev->impl_opaque = (uint8_t)DPAA_INVALID_MBUF_SEQN;
+	mbuf->seqn = DPAA_INVALID_MBUF_SEQN;
+	*bufs = mbuf;
+
+	return qman_cb_dqrr_consume;
+}
+
+enum qman_cb_dqrr_result
+dpaa_rx_cb_atomic(void *event,
+		  struct qman_portal *qm __always_unused,
+		  struct qman_fq *fq,
+		  const struct qm_dqrr_entry *dqrr,
+		  void **bufs)
+{
+	u8 index;
+	u32 ifid = ((struct dpaa_if *)fq->dpaa_intf)->ifid;
+	struct rte_mbuf *mbuf;
+	struct rte_event *ev = (struct rte_event *)event;
+
+	mbuf = dpaa_eth_fd_to_mbuf(&dqrr->fd, ifid);
+	ev->event_ptr = (void *)mbuf;
+	ev->flow_id = fq->ev.flow_id;
+	ev->sub_event_type = fq->ev.sub_event_type;
+	ev->event_type = RTE_EVENT_TYPE_ETHDEV;
+	ev->op = RTE_EVENT_OP_NEW;
+	ev->sched_type = fq->ev.sched_type;
+	ev->queue_id = fq->ev.queue_id;
+	ev->priority = fq->ev.priority;
+
+	/* Save active dqrr entries */
+	index = DQRR_PTR2IDX(dqrr);
+	DPAA_PER_LCORE_DQRR_SIZE++;
+	DPAA_PER_LCORE_DQRR_HELD |= 1 << index;
+	DPAA_PER_LCORE_DQRR_MBUF(index) = mbuf;
+	ev->impl_opaque = index + 1;
+	mbuf->seqn = (uint32_t)index + 1;
+	*bufs = mbuf;
+
+	return qman_cb_dqrr_defer;
+}
+
 uint16_t dpaa_eth_queue_rx(void *q,
 			   struct rte_mbuf **bufs,
 			   uint16_t nb_bufs)
@@ -432,19 +596,37 @@ uint16_t dpaa_eth_queue_rx(void *q,
 	struct qman_fq *fq = q;
 	struct qm_dqrr_entry *dq;
 	uint32_t num_rx = 0, ifid = ((struct dpaa_if *)fq->dpaa_intf)->ifid;
-	int ret;
+	int num_rx_bufs, ret;
+	uint32_t vdqcr_flags = 0;
+
+	if (unlikely(rte_dpaa_bpid_info == NULL &&
+				rte_eal_process_type() == RTE_PROC_SECONDARY))
+		rte_dpaa_bpid_info = fq->bp_array;
 
 	if (likely(fq->is_static))
 		return dpaa_eth_queue_portal_rx(fq, bufs, nb_bufs);
 
-	ret = rte_dpaa_portal_init((void *)0);
-	if (ret) {
-		DPAA_PMD_ERR("Failure in affining portal");
-		return 0;
+	if (unlikely(!RTE_PER_LCORE(dpaa_io))) {
+		ret = rte_dpaa_portal_init((void *)0);
+		if (ret) {
+			DPAA_PMD_ERR("Failure in affining portal");
+			return 0;
+		}
 	}
 
-	ret = qman_set_vdq(fq, (nb_bufs > DPAA_MAX_DEQUEUE_NUM_FRAMES) ?
-				DPAA_MAX_DEQUEUE_NUM_FRAMES : nb_bufs);
+	/* Until request for four buffers, we provide exact number of buffers.
+	 * Otherwise we do not set the QM_VDQCR_EXACT flag.
+	 * Not setting QM_VDQCR_EXACT flag can provide two more buffers than
+	 * requested, so we request two less in this case.
+	 */
+	if (nb_bufs < 4) {
+		vdqcr_flags = QM_VDQCR_EXACT;
+		num_rx_bufs = nb_bufs;
+	} else {
+		num_rx_bufs = nb_bufs > DPAA_MAX_DEQUEUE_NUM_FRAMES ?
+			(DPAA_MAX_DEQUEUE_NUM_FRAMES - 2) : (nb_bufs - 2);
+	}
+	ret = qman_set_vdq(fq, num_rx_bufs, vdqcr_flags);
 	if (ret)
 		return 0;
 
@@ -462,7 +644,7 @@ uint16_t dpaa_eth_queue_rx(void *q,
 static void *dpaa_get_pktbuf(struct dpaa_bp_info *bp_info)
 {
 	int ret;
-	uint64_t buf = 0;
+	size_t buf = 0;
 	struct bm_buffer bufs;
 
 	ret = bman_acquire(bp_info->bp, &bufs, 1, 0);
@@ -471,10 +653,11 @@ static void *dpaa_get_pktbuf(struct dpaa_bp_info *bp_info)
 		return (void *)buf;
 	}
 
-	DPAA_DP_LOG(DEBUG, "got buffer 0x%lx from pool %d",
+	DPAA_DP_LOG(DEBUG, "got buffer 0x%" PRIx64 " from pool %d",
 		    (uint64_t)bufs.addr, bufs.bpid);
 
-	buf = (uint64_t)rte_dpaa_mem_ptov(bufs.addr) - bp_info->meta_data_size;
+	buf = (size_t)DPAA_MEMPOOL_PTOV(bp_info, bufs.addr)
+				- bp_info->meta_data_size;
 	if (!buf)
 		goto out;
 
@@ -694,6 +877,8 @@ tx_on_external_pool(struct qman_fq *txq, struct rte_mbuf *mbuf,
 	}
 
 	DPAA_MBUF_TO_CONTIG_FD(dmable_mbuf, fd_arr, dpaa_intf->bp_info->bpid);
+	if (mbuf->ol_flags & DPAA_TX_CKSUM_OFFLOAD_MASK)
+		dpaa_unsegmented_checksum(mbuf, fd_arr);
 
 	return 0;
 }
@@ -708,11 +893,14 @@ dpaa_eth_queue_tx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
 	uint32_t frames_to_send, loop, sent = 0;
 	uint16_t state;
 	int ret;
+	uint32_t seqn, index, flags[DPAA_TX_BURST_SIZE] = {0};
 
-	ret = rte_dpaa_portal_init((void *)0);
-	if (ret) {
-		DPAA_PMD_ERR("Failure in affining portal");
-		return 0;
+	if (unlikely(!RTE_PER_LCORE(dpaa_io))) {
+		ret = rte_dpaa_portal_init((void *)0);
+		if (ret) {
+			DPAA_PMD_ERR("Failure in affining portal");
+			return 0;
+		}
 	}
 
 	DPAA_DP_LOG(DEBUG, "Transmitting %d buffers on queue: %p", nb_bufs, q);
@@ -722,6 +910,19 @@ dpaa_eth_queue_tx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
 				DPAA_TX_BURST_SIZE : nb_bufs;
 		for (loop = 0; loop < frames_to_send; loop++) {
 			mbuf = *(bufs++);
+			seqn = mbuf->seqn;
+			if (seqn != DPAA_INVALID_MBUF_SEQN) {
+				index = seqn - 1;
+				if (DPAA_PER_LCORE_DQRR_HELD & (1 << index)) {
+					flags[loop] =
+					   ((index & QM_EQCR_DCA_IDXMASK) << 8);
+					flags[loop] |= QMAN_ENQUEUE_FLAG_DCA;
+					DPAA_PER_LCORE_DQRR_SIZE--;
+					DPAA_PER_LCORE_DQRR_HELD &=
+								~(1 << index);
+				}
+			}
+
 			if (likely(RTE_MBUF_DIRECT(mbuf))) {
 				mp = mbuf->pool;
 				bp_info = DPAA_MEMPOOL_TO_POOL_INFO(mp);
@@ -756,6 +957,16 @@ dpaa_eth_queue_tx(void *q, struct rte_mbuf **bufs, uint16_t nb_bufs)
 					goto send_pkts;
 				}
 			} else {
+				/* TODO not supporting sg for external bufs*/
+				if (unlikely(mbuf->nb_segs > 1)) {
+					/* Set frames_to_send & nb_bufs so
+					 * that packets are transmitted till
+					 * previous frame.
+					 */
+					frames_to_send = loop;
+					nb_bufs = loop;
+					goto send_pkts;
+				}
 				state = tx_on_external_pool(q, mbuf,
 							    &fd_arr[loop]);
 				if (unlikely(state)) {
@@ -774,8 +985,8 @@ send_pkts:
 		loop = 0;
 		while (loop < frames_to_send) {
 			loop += qman_enqueue_multi(q, &fd_arr[loop],
-						   NULL,
-					frames_to_send - loop);
+						   &flags[loop],
+						   frames_to_send - loop);
 		}
 		nb_bufs -= frames_to_send;
 		sent += frames_to_send;