DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Wang, Zhihong" <zhihong.wang@intel.com>
To: Maxime Coquelin <maxime.coquelin@redhat.com>,
	"dev@dpdk.org" <dev@dpdk.org>
Cc: "yuanhan.liu@linux.intel.com" <yuanhan.liu@linux.intel.com>
Subject: Re: [dpdk-dev] [PATCH v3 1/5] vhost: rewrite enqueue
Date: Tue, 23 Aug 2016 02:27:15 +0000	[thread overview]
Message-ID: <8F6C2BD409508844A0EFC19955BE094110773477@SHSMSX103.ccr.corp.intel.com> (raw)
In-Reply-To: <11aee27f-3604-7706-ca6f-fcee77aa09cb@redhat.com>

Hi Maxime,

Thanks very much for the detailed review.

> -----Original Message-----
> From: Maxime Coquelin [mailto:maxime.coquelin@redhat.com]
> Sent: Monday, August 22, 2016 5:36 PM
> To: Wang, Zhihong <zhihong.wang@intel.com>; dev@dpdk.org
> Cc: yuanhan.liu@linux.intel.com
> Subject: Re: [PATCH v3 1/5] vhost: rewrite enqueue
> 
> 
> 
> On 08/19/2016 07:43 AM, Zhihong Wang wrote:
> > This patch implements the vhost logic from scratch into a single function
> > designed for high performance and better maintainability.
> >
> > ---
> > Changes in v3:
> >
> >  1. Rewrite enqueue and delete the obsolete in the same patch.
> >
> > Signed-off-by: Zhihong Wang <zhihong.wang@intel.com>
> > ---
> >  lib/librte_vhost/vhost_rxtx.c | 537 +++++++++++++-----------------------------
> >  1 file changed, 160 insertions(+), 377 deletions(-)
> >
> > diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost_rxtx.c
> > index 08a73fd..b09a9c3 100644
> > --- a/lib/librte_vhost/vhost_rxtx.c
> > +++ b/lib/librte_vhost/vhost_rxtx.c
> > @@ -91,7 +91,7 @@ is_valid_virt_queue_idx(uint32_t idx, int is_tx, uint32_t
> qp_nb)
> >  	return (is_tx ^ (idx & 1)) == 0 && idx < qp_nb * VIRTIO_QNUM;
> >  }
> >
> > -static void
> > +static inline void __attribute__((always_inline))
> >  virtio_enqueue_offload(struct rte_mbuf *m_buf, struct virtio_net_hdr
> *net_hdr)
> >  {
> >  	if (m_buf->ol_flags & PKT_TX_L4_MASK) {
> > @@ -125,427 +125,210 @@ virtio_enqueue_offload(struct rte_mbuf *m_buf,
> struct virtio_net_hdr *net_hdr)
> >  	}
> >  }
> >
> > -static inline void
> > -copy_virtio_net_hdr(struct virtio_net *dev, uint64_t desc_addr,
> > -		    struct virtio_net_hdr_mrg_rxbuf hdr)
> > +static inline uint32_t __attribute__((always_inline))
> > +loop_check(struct vhost_virtqueue *vq, uint16_t avail_idx, uint32_t pkt_left)
> Creating a function just for doing this doesn't make much sense.
> And the function name doesn't help.
> I think you should just remove this function.

Okay.

> 
> >  {
> > -	if (dev->vhost_hlen == sizeof(struct virtio_net_hdr_mrg_rxbuf))
> > -		*(struct virtio_net_hdr_mrg_rxbuf *)(uintptr_t)desc_addr = hdr;
> > -	else
> > -		*(struct virtio_net_hdr *)(uintptr_t)desc_addr = hdr.hdr;
> > +	if (pkt_left == 0 || avail_idx == vq->last_used_idx)
> > +		return 1;
> > +
> > +	return 0;
> >  }
> >
> > -static inline int __attribute__((always_inline))
> > -copy_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
> > -		  struct rte_mbuf *m, uint16_t desc_idx)
> > +static inline uint32_t __attribute__((always_inline))
> > +enqueue_packet(struct virtio_net *dev, struct vhost_virtqueue *vq,
> > +		uint16_t avail_idx, struct rte_mbuf *mbuf,
> > +		uint32_t is_mrg_rxbuf)
> >  {
> > -	uint32_t desc_avail, desc_offset;
> > -	uint32_t mbuf_avail, mbuf_offset;
> > -	uint32_t cpy_len;
> > +	struct virtio_net_hdr_mrg_rxbuf *virtio_hdr;
> >  	struct vring_desc *desc;
> > -	uint64_t desc_addr;
> > -	struct virtio_net_hdr_mrg_rxbuf virtio_hdr = {{0, 0, 0, 0, 0, 0}, 0};
> > -
> > -	desc = &vq->desc[desc_idx];
> > +	uint64_t desc_addr = 0;
> > +	uint32_t desc_chain_head = 0;
> > +	uint32_t desc_chain_len = 0;
> > +	uint32_t desc_current = 0;
> > +	uint32_t desc_offset = 0;
> > +	uint32_t mbuf_len = 0;
> > +	uint32_t mbuf_avail = 0;
> > +	uint32_t copy_len = 0;
> > +	uint32_t extra_buffers = 0;
> > +	uint32_t used_idx_round = 0;
> Most of these variables don't need to be initialized.

Okay.

> 
> > +
> > +	/* start with the first mbuf of the packet */
> > +	mbuf_len = rte_pktmbuf_data_len(mbuf);
> > +	mbuf_avail = mbuf_len;
> > +
> > +	/* get the current desc */
> > +	desc_current = vq->avail->ring[(vq->last_used_idx) & (vq->size - 1)];
> > +	desc_chain_head = desc_current;
> > +	desc = &vq->desc[desc_current];
> >  	desc_addr = gpa_to_vva(dev, desc->addr);
> > -	/*
> > -	 * Checking of 'desc_addr' placed outside of 'unlikely' macro to avoid
> > -	 * performance issue with some versions of gcc (4.8.4 and 5.3.0) which
> > -	 * otherwise stores offset on the stack instead of in a register.
> > -	 */
> > -	if (unlikely(desc->len < dev->vhost_hlen) || !desc_addr)
> > -		return -1;
> > -
> > -	rte_prefetch0((void *)(uintptr_t)desc_addr);
> > +	if (unlikely(!desc_addr))
> > +		goto error;
> >
> > -	virtio_enqueue_offload(m, &virtio_hdr.hdr);
> > -	copy_virtio_net_hdr(dev, desc_addr, virtio_hdr);
> > +	/* handle virtio header */
> > +	virtio_hdr = (struct virtio_net_hdr_mrg_rxbuf *)(uintptr_t)desc_addr;
> > +	virtio_enqueue_offload(mbuf, &(virtio_hdr->hdr));
> Parenthesis around virtio_hdr->hdr shouldn't be needed.
> >  	vhost_log_write(dev, desc->addr, dev->vhost_hlen);
> > -	PRINT_PACKET(dev, (uintptr_t)desc_addr, dev->vhost_hlen, 0);
> Looks like you remove the PRINT_PACKET calls.
> Does it impact performance?
> In any case, it should be mentionned in the commit message.

Will add this.

> 
> > -
> >  	desc_offset = dev->vhost_hlen;
> > -	desc_avail  = desc->len - dev->vhost_hlen;
> > +	desc_chain_len = desc_offset;
> > +	desc_addr += desc_offset;
> > +	if (is_mrg_rxbuf)
> > +		virtio_hdr->num_buffers = 1;
> >
> > -	mbuf_avail  = rte_pktmbuf_data_len(m);
> > -	mbuf_offset = 0;
> > -	while (mbuf_avail != 0 || m->next != NULL) {
> > -		/* done with current mbuf, fetch next */
> > -		if (mbuf_avail == 0) {
> > -			m = m->next;
> > -
> > -			mbuf_offset = 0;
> > -			mbuf_avail  = rte_pktmbuf_data_len(m);
> > +	/* start copy from mbuf to desc */
> > +	while (1) {
> Please avoid while(1) when you can check for a real condition:
> while (mbuf_avail || mbuf->next) ?

Will rewrite this logic.

> 
> Compiler should optimize this properly, no?
> > +		/* get the next mbuf if the current done */
> > +		if (!mbuf_avail) {
> > +			if (mbuf->next) {
> > +				mbuf = mbuf->next;
> > +				mbuf_len = rte_pktmbuf_data_len(mbuf);
> > +				mbuf_avail = mbuf_len;
> > +			} else
> > +				break;
> >  		}
> >
> > -		/* done with current desc buf, fetch next */
> > -		if (desc_avail == 0) {
> > -			if ((desc->flags & VRING_DESC_F_NEXT) == 0) {
> > -				/* Room in vring buffer is not enough */
> > -				return -1;
> > -			}
> > -			if (unlikely(desc->next >= vq->size))
> > -				return -1;
> > -
> > -			desc = &vq->desc[desc->next];
> > -			desc_addr = gpa_to_vva(dev, desc->addr);
> > -			if (unlikely(!desc_addr))
> > -				return -1;
> > -
> > -			desc_offset = 0;
> > -			desc_avail  = desc->len;
> > +		/* get the next desc if the current done */
> > +		if (desc->len <= desc_offset) {
> > +			if (desc->flags & VRING_DESC_F_NEXT) {
> > +				/* go on with the current desc chain */
> > +				desc_offset = 0;
> > +				desc_current = desc->next;
> > +				desc = &vq->desc[desc_current];
> > +				desc_addr = gpa_to_vva(dev, desc->addr);
> > +				if (unlikely(!desc_addr))
> > +					goto rollback;
> you could goto directly to error, and decrement last_used_idx directly
> under "error"'s goto since extra_buffers will be zero otherwise.

Good call.

> 
> Also, except desc_current affectation, all the above code is common
> with mergeable case, so you should avoid duplication.
> > +			} else if (is_mrg_rxbuf) {
> > +				/* start with the next desc chain */
> > +				used_idx_round = vq->last_used_idx
> > +					& (vq->size - 1);
> > +				vq->used->ring[used_idx_round].id =
> > +					desc_chain_head;
> > +				vq->used->ring[used_idx_round].len =
> > +					desc_chain_len;
> > +				vhost_log_used_vring(dev, vq,
> > +					offsetof(struct vring_used,
> > +						ring[used_idx_round]),
> > +					sizeof(vq->used->ring[
> > +						used_idx_round]));
> > +				vq->last_used_idx++;
> > +				extra_buffers++;
> > +				virtio_hdr->num_buffers++;
> > +				if (avail_idx == vq->last_used_idx)
> > +					goto rollback;
> > +
> > +				desc_current =
> > +					vq->avail->ring[(vq->last_used_idx) &
> > +					(vq->size - 1)];
> > +				desc_chain_head = desc_current;
> > +				desc = &vq->desc[desc_current];
> > +				desc_addr = gpa_to_vva(dev, desc->addr);
> > +				if (unlikely(!desc_addr))
> > +					goto rollback;
> > +
> > +				desc_chain_len = 0;
> > +				desc_offset = 0;
> > +			} else
> > +				goto rollback;
> >  		}
> >
> > -		cpy_len = RTE_MIN(desc_avail, mbuf_avail);
> > -		rte_memcpy((void *)((uintptr_t)(desc_addr + desc_offset)),
> > -			rte_pktmbuf_mtod_offset(m, void *, mbuf_offset),
> > -			cpy_len);
> > -		vhost_log_write(dev, desc->addr + desc_offset, cpy_len);
> > -		PRINT_PACKET(dev, (uintptr_t)(desc_addr + desc_offset),
> > -			     cpy_len, 0);
> > -
> > -		mbuf_avail  -= cpy_len;
> > -		mbuf_offset += cpy_len;
> > -		desc_avail  -= cpy_len;
> > -		desc_offset += cpy_len;
> > -	}
> > -
> > -	return 0;
> > -}
> > -
> > -/**
> > - * This function adds buffers to the virtio devices RX virtqueue. Buffers can
> > - * be received from the physical port or from another virtio device. A packet
> > - * count is returned to indicate the number of packets that are succesfully
> > - * added to the RX queue. This function works when the mbuf is scattered, but
> > - * it doesn't support the mergeable feature.
> > - */
> > -static inline uint32_t __attribute__((always_inline))
> > -virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
> > -	      struct rte_mbuf **pkts, uint32_t count)
> > -{
> > -	struct vhost_virtqueue *vq;
> > -	uint16_t avail_idx, free_entries, start_idx;
> > -	uint16_t desc_indexes[MAX_PKT_BURST];
> > -	uint16_t used_idx;
> > -	uint32_t i;
> > -
> > -	LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
> > -	if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->virt_qp_nb))) {
> > -		RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
> > -			dev->vid, __func__, queue_id);
> > -		return 0;
> > +		/* copy mbuf data */
> > +		copy_len = RTE_MIN(desc->len - desc_offset, mbuf_avail);
> > +		rte_memcpy((void *)(uintptr_t)desc_addr,
> > +				rte_pktmbuf_mtod_offset(mbuf, void *,
> > +					mbuf_len - mbuf_avail),
> > +				copy_len);
> > +		vhost_log_write(dev, desc->addr + desc_offset, copy_len);
> > +		mbuf_avail -= copy_len;
> > +		desc_offset += copy_len;
> > +		desc_addr += copy_len;
> > +		desc_chain_len += copy_len;
> >  	}
> >
> > -	vq = dev->virtqueue[queue_id];
> > -	if (unlikely(vq->enabled == 0))
> > -		return 0;
> > -
> > -	avail_idx = *((volatile uint16_t *)&vq->avail->idx);
> > -	start_idx = vq->last_used_idx;
> > -	free_entries = avail_idx - start_idx;
> > -	count = RTE_MIN(count, free_entries);
> > -	count = RTE_MIN(count, (uint32_t)MAX_PKT_BURST);
> > -	if (count == 0)
> > -		return 0;
> > -
> > -	LOG_DEBUG(VHOST_DATA, "(%d) start_idx %d | end_idx %d\n",
> > -		dev->vid, start_idx, start_idx + count);
> > -
> > -	/* Retrieve all of the desc indexes first to avoid caching issues. */
> > -	rte_prefetch0(&vq->avail->ring[start_idx & (vq->size - 1)]);
> > -	for (i = 0; i < count; i++) {
> > -		used_idx = (start_idx + i) & (vq->size - 1);
> > -		desc_indexes[i] = vq->avail->ring[used_idx];
> > -		vq->used->ring[used_idx].id = desc_indexes[i];
> > -		vq->used->ring[used_idx].len = pkts[i]->pkt_len +
> > -					       dev->vhost_hlen;
> > -		vhost_log_used_vring(dev, vq,
> > -			offsetof(struct vring_used, ring[used_idx]),
> > -			sizeof(vq->used->ring[used_idx]));
> > -	}
> > +	used_idx_round = vq->last_used_idx & (vq->size - 1);
> > +	vq->used->ring[used_idx_round].id = desc_chain_head;
> > +	vq->used->ring[used_idx_round].len = desc_chain_len;
> > +	vhost_log_used_vring(dev, vq,
> > +		offsetof(struct vring_used, ring[used_idx_round]),
> > +		sizeof(vq->used->ring[used_idx_round]));
> > +	vq->last_used_idx++;
> All this code is duplicatedd from the rx_mergeable base.
> I think a dedicated inline function would really make sense here.

Good catch. Will make a function for this.

> 
> >
> > -	rte_prefetch0(&vq->desc[desc_indexes[0]]);
> > -	for (i = 0; i < count; i++) {
> > -		uint16_t desc_idx = desc_indexes[i];
> > -		int err;
> > +	return 0;
> >
> > -		err = copy_mbuf_to_desc(dev, vq, pkts[i], desc_idx);
> > -		if (unlikely(err)) {
> > -			used_idx = (start_idx + i) & (vq->size - 1);
> > -			vq->used->ring[used_idx].len = dev->vhost_hlen;
> > -			vhost_log_used_vring(dev, vq,
> > -				offsetof(struct vring_used, ring[used_idx]),
> > -				sizeof(vq->used->ring[used_idx]));
> > -		}
> > +rollback:
> > +	/* rollback on any error if last_used_idx update on-the-fly */
> > +	if (is_mrg_rxbuf)
> If (!is_mrg_rxbuf), extra_buffers will be zero, so just remove the test,
> and place belw line directly under error: as explained above.

Sure. Thanks.

> > +		vq->last_used_idx -= extra_buffers;
> >
> > -		if (i + 1 < count)
> > -			rte_prefetch0(&vq->desc[desc_indexes[i+1]]);
> > -	}
> > +error:
> > +	return 1;
> > +}
> >
> > +static inline void __attribute__((always_inline))
> > +notify_guest(struct virtio_net *dev, struct vhost_virtqueue *vq)
> > +{
> >  	rte_smp_wmb();
> > -
> > -	*(volatile uint16_t *)&vq->used->idx += count;
> > -	vq->last_used_idx += count;
> > -	vhost_log_used_vring(dev, vq,
> > -		offsetof(struct vring_used, idx),
> > -		sizeof(vq->used->idx));
> > -
> > -	/* flush used->idx update before we read avail->flags. */
> > +	vq->used->idx = vq->last_used_idx;
> > +	vhost_log_used_vring(dev, vq, offsetof(struct vring_used, idx),
> > +			sizeof(vq->used->idx));
> >  	rte_mb();
> > -
> > -	/* Kick the guest if necessary. */
> >  	if (!(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
> >  			&& (vq->callfd >= 0))
> >  		eventfd_write(vq->callfd, (eventfd_t)1);
> > -	return count;
> > -}
> > -
> > -static inline int
> > -fill_vec_buf(struct vhost_virtqueue *vq, uint32_t avail_idx,
> > -	     uint32_t *allocated, uint32_t *vec_idx,
> > -	     struct buf_vector *buf_vec)
> > -{
> > -	uint16_t idx = vq->avail->ring[avail_idx & (vq->size - 1)];
> > -	uint32_t vec_id = *vec_idx;
> > -	uint32_t len    = *allocated;
> > -
> > -	while (1) {
> > -		if (unlikely(vec_id >= BUF_VECTOR_MAX || idx >= vq->size))
> > -			return -1;
> > -
> > -		len += vq->desc[idx].len;
> > -		buf_vec[vec_id].buf_addr = vq->desc[idx].addr;
> > -		buf_vec[vec_id].buf_len  = vq->desc[idx].len;
> > -		buf_vec[vec_id].desc_idx = idx;
> > -		vec_id++;
> > -
> > -		if ((vq->desc[idx].flags & VRING_DESC_F_NEXT) == 0)
> > -			break;
> > -
> > -		idx = vq->desc[idx].next;
> > -	}
> > -
> > -	*allocated = len;
> > -	*vec_idx   = vec_id;
> > -
> > -	return 0;
> > -}
> > -
> > -/*
> > - * Returns -1 on fail, 0 on success
> > - */
> > -static inline int
> > -reserve_avail_buf_mergeable(struct vhost_virtqueue *vq, uint32_t size,
> > -			    uint16_t *end, struct buf_vector *buf_vec)
> > -{
> > -	uint16_t cur_idx;
> > -	uint16_t avail_idx;
> > -	uint32_t allocated = 0;
> > -	uint32_t vec_idx = 0;
> > -	uint16_t tries = 0;
> > -
> > -	cur_idx  = vq->last_used_idx;
> > -
> > -	while (1) {
> > -		avail_idx = *((volatile uint16_t *)&vq->avail->idx);
> > -		if (unlikely(cur_idx == avail_idx))
> > -			return -1;
> > -
> > -		if (unlikely(fill_vec_buf(vq, cur_idx, &allocated,
> > -					  &vec_idx, buf_vec) < 0))
> > -			return -1;
> > -
> > -		cur_idx++;
> > -		tries++;
> > -
> > -		if (allocated >= size)
> > -			break;
> > -
> > -		/*
> > -		 * if we tried all available ring items, and still
> > -		 * can't get enough buf, it means something abnormal
> > -		 * happened.
> > -		 */
> > -		if (unlikely(tries >= vq->size))
> > -			return -1;
> > -	}
> > -
> > -	*end = cur_idx;
> > -	return 0;
> >  }
> >
> > -static inline uint32_t __attribute__((always_inline))
> > -copy_mbuf_to_desc_mergeable(struct virtio_net *dev, struct
> vhost_virtqueue *vq,
> > -			    uint16_t end_idx, struct rte_mbuf *m,
> > -			    struct buf_vector *buf_vec)
> > +uint16_t
> > +rte_vhost_enqueue_burst(int vid, uint16_t queue_id,
> > +	struct rte_mbuf **pkts, uint16_t count)
> >  {
> > -	struct virtio_net_hdr_mrg_rxbuf virtio_hdr = {{0, 0, 0, 0, 0, 0}, 0};
> > -	uint32_t vec_idx = 0;
> > -	uint16_t start_idx = vq->last_used_idx;
> > -	uint16_t cur_idx = start_idx;
> > -	uint64_t desc_addr;
> > -	uint32_t mbuf_offset, mbuf_avail;
> > -	uint32_t desc_offset, desc_avail;
> > -	uint32_t cpy_len;
> > -	uint16_t desc_idx, used_idx;
> > -
> > -	if (unlikely(m == NULL))
> > +	struct vhost_virtqueue *vq;
> > +	struct virtio_net *dev;
> > +	uint32_t pkt_idx = 0;
> > +	uint32_t pkt_left = 0;
> > +	uint32_t pkt_sent = 0;
> > +	uint32_t is_mrg_rxbuf = 0;
> > +	uint16_t avail_idx = 0;
> > +
> > +	/* precheck */
> Comment not very informative here.

Okay.

> > +	if (unlikely(count == 0))
> >  		return 0;
> >
> > -	LOG_DEBUG(VHOST_DATA, "(%d) current index %d | end index %d\n",
> > -		dev->vid, cur_idx, end_idx);
> > +	count = RTE_MIN((uint32_t)MAX_PKT_BURST, count);
> >
> > -	desc_addr = gpa_to_vva(dev, buf_vec[vec_idx].buf_addr);
> > -	if (buf_vec[vec_idx].buf_len < dev->vhost_hlen || !desc_addr)
> > +	dev = get_device(vid);
> > +	if (unlikely(!dev))
> >  		return 0;
> >
> > -	rte_prefetch0((void *)(uintptr_t)desc_addr);
> > -
> > -	virtio_hdr.num_buffers = end_idx - start_idx;
> > -	LOG_DEBUG(VHOST_DATA, "(%d) RX: num merge buffers %d\n",
> > -		dev->vid, virtio_hdr.num_buffers);
> > -
> > -	virtio_enqueue_offload(m, &virtio_hdr.hdr);
> > -	copy_virtio_net_hdr(dev, desc_addr, virtio_hdr);
> > -	vhost_log_write(dev, buf_vec[vec_idx].buf_addr, dev->vhost_hlen);
> > -	PRINT_PACKET(dev, (uintptr_t)desc_addr, dev->vhost_hlen, 0);
> > -
> > -	desc_avail  = buf_vec[vec_idx].buf_len - dev->vhost_hlen;
> > -	desc_offset = dev->vhost_hlen;
> > -
> > -	mbuf_avail  = rte_pktmbuf_data_len(m);
> > -	mbuf_offset = 0;
> > -	while (mbuf_avail != 0 || m->next != NULL) {
> > -		/* done with current desc buf, get the next one */
> > -		if (desc_avail == 0) {
> > -			desc_idx = buf_vec[vec_idx].desc_idx;
> > -
> > -			if (!(vq->desc[desc_idx].flags & VRING_DESC_F_NEXT)) {
> > -				/* Update used ring with desc information */
> > -				used_idx = cur_idx++ & (vq->size - 1);
> > -				vq->used->ring[used_idx].id  = desc_idx;
> > -				vq->used->ring[used_idx].len = desc_offset;
> > -				vhost_log_used_vring(dev, vq,
> > -					offsetof(struct vring_used,
> > -						 ring[used_idx]),
> > -					sizeof(vq->used->ring[used_idx]));
> > -			}
> > -
> > -			vec_idx++;
> > -			desc_addr = gpa_to_vva(dev, buf_vec[vec_idx].buf_addr);
> > -			if (unlikely(!desc_addr))
> > -				return 0;
> > -
> > -			/* Prefetch buffer address. */
> > -			rte_prefetch0((void *)(uintptr_t)desc_addr);
> > -			desc_offset = 0;
> > -			desc_avail  = buf_vec[vec_idx].buf_len;
> > -		}
> > -
> > -		/* done with current mbuf, get the next one */
> > -		if (mbuf_avail == 0) {
> > -			m = m->next;
> > -
> > -			mbuf_offset = 0;
> > -			mbuf_avail  = rte_pktmbuf_data_len(m);
> > -		}
> > -
> > -		cpy_len = RTE_MIN(desc_avail, mbuf_avail);
> > -		rte_memcpy((void *)((uintptr_t)(desc_addr + desc_offset)),
> > -			rte_pktmbuf_mtod_offset(m, void *, mbuf_offset),
> > -			cpy_len);
> > -		vhost_log_write(dev, buf_vec[vec_idx].buf_addr + desc_offset,
> > -			cpy_len);
> > -		PRINT_PACKET(dev, (uintptr_t)(desc_addr + desc_offset),
> > -			cpy_len, 0);
> > -
> > -		mbuf_avail  -= cpy_len;
> > -		mbuf_offset += cpy_len;
> > -		desc_avail  -= cpy_len;
> > -		desc_offset += cpy_len;
> > -	}
> > -
> > -	used_idx = cur_idx & (vq->size - 1);
> > -	vq->used->ring[used_idx].id = buf_vec[vec_idx].desc_idx;
> > -	vq->used->ring[used_idx].len = desc_offset;
> > -	vhost_log_used_vring(dev, vq,
> > -		offsetof(struct vring_used, ring[used_idx]),
> > -		sizeof(vq->used->ring[used_idx]));
> > -
> > -	return end_idx - start_idx;
> > -}
> > -
> > -static inline uint32_t __attribute__((always_inline))
> > -virtio_dev_merge_rx(struct virtio_net *dev, uint16_t queue_id,
> > -	struct rte_mbuf **pkts, uint32_t count)
> > -{
> > -	struct vhost_virtqueue *vq;
> > -	uint32_t pkt_idx = 0, nr_used = 0;
> > -	uint16_t end;
> > -	struct buf_vector buf_vec[BUF_VECTOR_MAX];
> > -
> > -	LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
> > -	if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->virt_qp_nb))) {
> > -		RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
> > -			dev->vid, __func__, queue_id);
> > +	if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->virt_qp_nb)))
> >  		return 0;
> > -	}
> >
> >  	vq = dev->virtqueue[queue_id];
> > -	if (unlikely(vq->enabled == 0))
> > +	if (unlikely(!vq->enabled))
> >  		return 0;
> >
> > -	count = RTE_MIN((uint32_t)MAX_PKT_BURST, count);
> > -	if (count == 0)
> > -		return 0;
> > +	if (dev->features & (1ULL << VIRTIO_NET_F_MRG_RXBUF))
> > +		is_mrg_rxbuf = 1;
> >
> > -	for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
> > -		uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
> > -
> > -		if (unlikely(reserve_avail_buf_mergeable(vq, pkt_len,
> > -							 &end, buf_vec) < 0)) {
> > -			LOG_DEBUG(VHOST_DATA,
> > -				"(%d) failed to get enough desc from vring\n",
> > -				dev->vid);
> > +	/* start enqueuing packets 1 by 1 */
> > +	pkt_idx = 0;
> > +	pkt_left = count;
> > +	avail_idx = *((volatile uint16_t *)&vq->avail->idx);
> > +	while (1) {
> > +		if (loop_check(vq, avail_idx, pkt_left))
> What about:
> while (pkt_left && avail_idx != vq->last_used_idx) {

Will rewrite it.

> 
> >  			break;
> > -		}
> > -
> > -		nr_used = copy_mbuf_to_desc_mergeable(dev, vq, end,
> > -						      pkts[pkt_idx], buf_vec);
> > -		rte_smp_wmb();
> >
> > -		*(volatile uint16_t *)&vq->used->idx += nr_used;
> > -		vhost_log_used_vring(dev, vq, offsetof(struct vring_used, idx),
> > -			sizeof(vq->used->idx));
> > -		vq->last_used_idx += nr_used;
> > -	}
> > -
> > -	if (likely(pkt_idx)) {
> > -		/* flush used->idx update before we read avail->flags. */
> > -		rte_mb();
> > +		if (enqueue_packet(dev, vq, avail_idx, pkts[pkt_idx],
> > +					is_mrg_rxbuf))
> > +			break;
> >
> > -		/* Kick the guest if necessary. */
> > -		if (!(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
> > -				&& (vq->callfd >= 0))
> > -			eventfd_write(vq->callfd, (eventfd_t)1);
> > +		pkt_idx++;
> > +		pkt_sent++;
> > +		pkt_left--;
> >  	}
> >
> > -	return pkt_idx;
> > -}
> > -
> > -uint16_t
> > -rte_vhost_enqueue_burst(int vid, uint16_t queue_id,
> > -	struct rte_mbuf **pkts, uint16_t count)
> > -{
> > -	struct virtio_net *dev = get_device(vid);
> > -
> > -	if (!dev)
> > -		return 0;
> > +	/* update used idx and kick the guest if necessary */
> > +	if (pkt_sent)
> > +		notify_guest(dev, vq);
> >
> > -	if (dev->features & (1 << VIRTIO_NET_F_MRG_RXBUF))
> > -		return virtio_dev_merge_rx(dev, queue_id, pkts, count);
> > -	else
> > -		return virtio_dev_rx(dev, queue_id, pkts, count);
> > +	return pkt_sent;
> >  }
> >
> >  static void
> >

  reply	other threads:[~2016-08-23  2:27 UTC|newest]

Thread overview: 140+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-16  3:50 [dpdk-dev] [PATCH] optimize vhost enqueue Zhihong Wang
2016-08-16 13:59 ` Maxime Coquelin
2016-08-17  1:45   ` Wang, Zhihong
2016-08-17  2:38     ` Yuanhan Liu
2016-08-17  6:41       ` Wang, Zhihong
2016-08-17  9:17         ` Maxime Coquelin
2016-08-17  9:51           ` Yuanhan Liu
2016-08-18 13:44             ` Wang, Zhihong
2016-08-17 10:07           ` Wang, Zhihong
2016-08-18  6:33 ` [dpdk-dev] [PATCH v2 0/6] vhost: optimize enqueue Zhihong Wang
2016-08-18  6:33   ` [dpdk-dev] [PATCH v2 1/6] vhost: rewrite enqueue Zhihong Wang
2016-08-19  2:39     ` Yuanhan Liu
2016-08-19  7:07       ` Wang, Zhihong
2016-08-18  6:33   ` [dpdk-dev] [PATCH v2 2/6] vhost: remove obsolete Zhihong Wang
2016-08-19  2:32     ` Yuanhan Liu
2016-08-19  7:08       ` Wang, Zhihong
2016-08-18  6:33   ` [dpdk-dev] [PATCH v2 3/6] vhost: remove useless volatile Zhihong Wang
2016-08-18  6:33   ` [dpdk-dev] [PATCH v2 4/6] vhost: add desc prefetch Zhihong Wang
2016-08-18  6:33   ` [dpdk-dev] [PATCH v2 5/6] vhost: batch update used ring Zhihong Wang
2016-08-18  6:33   ` [dpdk-dev] [PATCH v2 6/6] vhost: optimize cache access Zhihong Wang
2016-08-19  5:43 ` [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue Zhihong Wang
2016-08-19  5:43   ` [dpdk-dev] [PATCH v3 1/5] vhost: rewrite enqueue Zhihong Wang
2016-08-22  9:35     ` Maxime Coquelin
2016-08-23  2:27       ` Wang, Zhihong [this message]
2016-08-25  4:00       ` Yuanhan Liu
2016-08-19  5:43   ` [dpdk-dev] [PATCH v3 2/5] vhost: remove useless volatile Zhihong Wang
2016-08-19  5:43   ` [dpdk-dev] [PATCH v3 3/5] vhost: add desc prefetch Zhihong Wang
2016-08-19  5:43   ` [dpdk-dev] [PATCH v3 4/5] vhost: batch update used ring Zhihong Wang
2016-08-25  3:48     ` Yuanhan Liu
2016-08-25  5:19       ` Wang, Zhihong
2016-08-19  5:43   ` [dpdk-dev] [PATCH v3 5/5] vhost: optimize cache access Zhihong Wang
2016-08-22  8:11   ` [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue Maxime Coquelin
2016-08-22 10:01     ` Maxime Coquelin
2016-08-22 10:35       ` Thomas Monjalon
2016-08-24  3:37         ` Wang, Zhihong
2016-08-23  2:31       ` Wang, Zhihong
2016-08-23 10:43         ` Wang, Zhihong
2016-08-23 12:16           ` Maxime Coquelin
2016-08-23 12:22           ` Yuanhan Liu
2016-08-23  2:15     ` Wang, Zhihong
2016-09-21  8:50     ` Jianbo Liu
2016-09-21  9:27       ` Wang, Zhihong
2016-09-21 12:54         ` Jianbo Liu
2016-09-22  2:11           ` Wang, Zhihong
2016-09-22  2:29           ` Yuanhan Liu
2016-09-22  5:47             ` Jianbo Liu
2016-09-22  6:58               ` Wang, Zhihong
2016-09-22  9:01                 ` Jianbo Liu
2016-09-22 10:04                   ` Wang, Zhihong
2016-09-22 14:41                     ` Jianbo Liu
2016-09-23  2:56                       ` Wang, Zhihong
2016-09-23 10:41                         ` Jianbo Liu
2016-09-23 13:41                           ` Thomas Monjalon
2016-09-25  5:41                             ` Wang, Zhihong
2016-09-26  5:12                               ` Jianbo Liu
2016-09-26  5:25                                 ` Wang, Zhihong
2016-09-26  5:38                                   ` Jianbo Liu
2016-09-26  6:00                                     ` Wang, Zhihong
2016-09-26  4:24                             ` Jianbo Liu
2016-09-26  5:37                   ` Luke Gorrie
2016-09-26  5:40                     ` Jianbo Liu
2016-09-27 10:21                   ` Yuanhan Liu
2016-09-27 16:45                     ` Wang, Zhihong
2016-10-09 12:09                       ` Wang, Zhihong
2016-10-10  2:44                         ` Yuanhan Liu
2016-10-10  5:31                           ` Jianbo Liu
2016-10-10  6:22                             ` Wang, Zhihong
2016-10-10  6:57                               ` Jianbo Liu
2016-10-10  7:25                                 ` Wang, Zhihong
2016-10-12  2:53               ` Yuanhan Liu
2016-10-12 12:22                 ` Wang, Zhihong
2016-10-12 15:31                   ` Thomas Monjalon
2016-10-13  1:21                     ` Wang, Zhihong
2016-10-13  3:51                     ` Jianbo Liu
2016-10-13  5:33                   ` Yuanhan Liu
2016-10-13  5:35                     ` Yuanhan Liu
2016-10-13  6:02                     ` Wang, Zhihong
2016-10-13  7:54                       ` Maxime Coquelin
2016-10-13  9:23                         ` Maxime Coquelin
2016-10-14 10:11                           ` Yuanhan Liu
2016-08-30  3:35 ` [dpdk-dev] [PATCH v4 0/6] " Zhihong Wang
2016-08-30  3:35   ` [dpdk-dev] [PATCH v4 1/6] vhost: fix windows vm hang Zhihong Wang
2016-09-05  5:24     ` [dpdk-dev] [dpdk-stable] " Yuanhan Liu
2016-09-05  5:25       ` Wang, Zhihong
2016-09-05  5:40         ` Yuanhan Liu
2016-08-30  3:36   ` [dpdk-dev] [PATCH v4 2/6] vhost: rewrite enqueue Zhihong Wang
2016-09-05  6:39     ` Yuanhan Liu
2016-09-07  5:33       ` Yuanhan Liu
2016-09-07  5:39         ` Wang, Zhihong
2016-08-30  3:36   ` [dpdk-dev] [PATCH v4 3/6] vhost: remove useless volatile Zhihong Wang
2016-08-30  3:36   ` [dpdk-dev] [PATCH v4 4/6] vhost: add desc prefetch Zhihong Wang
2016-08-30  3:36   ` [dpdk-dev] [PATCH v4 5/6] vhost: batch update used ring Zhihong Wang
2016-08-30  3:36   ` [dpdk-dev] [PATCH v4 6/6] vhost: optimize cache access Zhihong Wang
2016-09-09  3:39 ` [dpdk-dev] [PATCH v5 0/6] vhost: optimize enqueue Zhihong Wang
2016-09-09  3:39   ` [dpdk-dev] [PATCH v5 1/6] vhost: fix windows vm hang Zhihong Wang
2016-09-09  3:39   ` [dpdk-dev] [PATCH v5 2/6] vhost: rewrite enqueue Zhihong Wang
2016-09-12 15:42     ` Maxime Coquelin
2016-09-14  8:20       ` Wang, Zhihong
2016-09-15 16:35         ` Maxime Coquelin
2016-09-12 16:26     ` Maxime Coquelin
2016-09-14  8:22       ` Wang, Zhihong
2016-09-18 14:19     ` Yuanhan Liu
2016-09-19  3:29       ` Wang, Zhihong
2016-09-09  3:39   ` [dpdk-dev] [PATCH v5 3/6] vhost: remove useless volatile Zhihong Wang
2016-09-09  3:39   ` [dpdk-dev] [PATCH v5 4/6] vhost: add desc prefetch Zhihong Wang
2016-09-09  3:39   ` [dpdk-dev] [PATCH v5 5/6] vhost: batch update used ring Zhihong Wang
2016-09-12 15:45     ` Maxime Coquelin
2016-09-14  8:43       ` Wang, Zhihong
2016-09-15 16:38         ` Maxime Coquelin
2016-09-18  2:55           ` Yuanhan Liu
2016-09-18  2:57             ` Wang, Zhihong
2016-09-09  3:39   ` [dpdk-dev] [PATCH v5 6/6] vhost: optimize cache access Zhihong Wang
2016-09-12 13:52   ` [dpdk-dev] [PATCH v5 0/6] vhost: optimize enqueue Maxime Coquelin
2016-09-12 13:56     ` Maxime Coquelin
2016-09-12 14:01     ` Yuanhan Liu
2016-09-20  2:00 ` [dpdk-dev] [PATCH v6 " Zhihong Wang
2016-09-20  2:00   ` [dpdk-dev] [PATCH v6 1/6] vhost: fix windows vm hang Zhihong Wang
2016-10-13  6:18     ` [dpdk-dev] [dpdk-stable] " Yuanhan Liu
2016-09-20  2:00   ` [dpdk-dev] [PATCH v6 2/6] vhost: rewrite enqueue Zhihong Wang
2016-09-22  9:58     ` Jianbo Liu
2016-09-22 10:13       ` Wang, Zhihong
2016-09-20  2:00   ` [dpdk-dev] [PATCH v6 3/6] vhost: remove useless volatile Zhihong Wang
2016-09-20  2:00   ` [dpdk-dev] [PATCH v6 4/6] vhost: add desc prefetch Zhihong Wang
2016-09-20  2:00   ` [dpdk-dev] [PATCH v6 5/6] vhost: batch update used ring Zhihong Wang
2016-09-20  2:00   ` [dpdk-dev] [PATCH v6 6/6] vhost: optimize cache access Zhihong Wang
2016-09-21  2:26   ` [dpdk-dev] [PATCH v6 0/6] vhost: optimize enqueue Yuanhan Liu
2016-09-21  4:39     ` Maxime Coquelin
2016-10-14  9:34   ` [dpdk-dev] [PATCH v7 0/7] vhost: optimize mergeable Rx path Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 1/7] vhost: remove useless volatile Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 2/7] vhost: optimize cache access Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 3/7] vhost: simplify mergeable Rx vring reservation Yuanhan Liu
2016-10-25 22:08       ` Thomas Monjalon
2016-10-26  2:56         ` Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 4/7] vhost: use last avail idx for avail ring reservation Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 5/7] vhost: shadow used ring update Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 6/7] vhost: prefetch avail ring Yuanhan Liu
2016-10-14  9:34     ` [dpdk-dev] [PATCH v7 7/7] vhost: retrieve avail head once Yuanhan Liu
2016-10-18  2:25     ` [dpdk-dev] [PATCH v7 0/7] vhost: optimize mergeable Rx path Jianbo Liu
2016-10-18 14:53     ` Maxime Coquelin
2016-10-21  7:51     ` Yuanhan Liu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8F6C2BD409508844A0EFC19955BE094110773477@SHSMSX103.ccr.corp.intel.com \
    --to=zhihong.wang@intel.com \
    --cc=dev@dpdk.org \
    --cc=maxime.coquelin@redhat.com \
    --cc=yuanhan.liu@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).