Re: [PATCH net-next 2/5] virtio_net: Add page_pool support to improve performance

From: Liang Chen
Date: Sat May 27 2023 - 08:35:23 EST

Next message: Arnaldo Carvalho de Melo: "Re: [PATCH v1] perf test python: Put perf python at start of sys.path"
Previous message: Liang Chen: "Re: [PATCH net-next 1/5] virtio_net: Fix an unsafe reference to the page chain"
In reply to: Jason Wang: "Re: [PATCH net-next 2/5] virtio_net: Add page_pool support to improve performance"
Next in thread: Michael S. Tsirkin: "Re: [PATCH net-next 2/5] virtio_net: Add page_pool support to improve performance"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Fri, May 26, 2023 at 2:51 PM Jason Wang <jasowang@xxxxxxxxxx> wrote:
>
> On Fri, May 26, 2023 at 1:46 PM Liang Chen <liangchen.linux@xxxxxxxxx> wrote:
> >
> > The implementation at the moment uses one page per packet in both the
> > normal and XDP path.
>
> It's better to explain why we need a page pool and how it can help the
> performance.
>

Sure, I will include that on v2.
> > In addition, introducing a module parameter to enable
> > or disable the usage of page pool (disabled by default).
>
> If page pool wins for most of the cases, any reason to disable it by default?
>

Thank you for raising the point. It does make sense to enable it by default.
> >
> > In single-core vm testing environments, it gives a modest performance gain
> > in the normal path.
> > Upstream codebase: 47.5 Gbits/sec
> > Upstream codebase + page_pool support: 50.2 Gbits/sec
> >
> > In multi-core vm testing environments, The most significant performance
> > gain is observed in XDP cpumap:
> > Upstream codebase: 1.38 Gbits/sec
> > Upstream codebase + page_pool support: 9.74 Gbits/sec
>
> Please show more details on the test. E.g which kinds of tests have
> you measured?
>
> Btw, it would be better to measure PPS as well.
>

Sure. It will be added on v2.
> >
> > With this foundation, we can further integrate page pool fragmentation and
> > DMA map/unmap support.
> >
> > Signed-off-by: Liang Chen <liangchen.linux@xxxxxxxxx>
> > ---
> > drivers/net/virtio_net.c | 188 ++++++++++++++++++++++++++++++---------
>
> I believe we should make virtio-net to select CONFIG_PAGE_POOL or do
> the ifdef tricks at least.
>

Sure. it will be done on v2.
> > 1 file changed, 146 insertions(+), 42 deletions(-)
> >
> > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > index c5dca0d92e64..99c0ca0c1781 100644
> > --- a/drivers/net/virtio_net.c
> > +++ b/drivers/net/virtio_net.c
> > @@ -31,6 +31,9 @@ module_param(csum, bool, 0444);
> > module_param(gso, bool, 0444);
> > module_param(napi_tx, bool, 0644);
> >
> > +static bool page_pool_enabled;
> > +module_param(page_pool_enabled, bool, 0400);
> > +
> > /* FIXME: MTU in config. */
> > #define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
> > #define GOOD_COPY_LEN 128
> > @@ -159,6 +162,9 @@ struct receive_queue {
> > /* Chain pages by the private ptr. */
> > struct page *pages;
> >
> > + /* Page pool */
> > + struct page_pool *page_pool;
> > +
> > /* Average packet length for mergeable receive buffers. */
> > struct ewma_pkt_len mrg_avg_pkt_len;
> >
> > @@ -459,6 +465,14 @@ static struct sk_buff *virtnet_build_skb(void *buf, unsigned int buflen,
> > return skb;
> > }
> >
> > +static void virtnet_put_page(struct receive_queue *rq, struct page *page)
> > +{
> > + if (rq->page_pool)
> > + page_pool_put_full_page(rq->page_pool, page, true);
> > + else
> > + put_page(page);
> > +}
> > +
> > /* Called from bottom half context */
> > static struct sk_buff *page_to_skb(struct virtnet_info *vi,
> > struct receive_queue *rq,
> > @@ -555,7 +569,7 @@ static struct sk_buff *page_to_skb(struct virtnet_info *vi,
> > hdr = skb_vnet_hdr(skb);
> > memcpy(hdr, hdr_p, hdr_len);
> > if (page_to_free)
> > - put_page(page_to_free);
> > + virtnet_put_page(rq, page_to_free);
> >
> > return skb;
> > }
> > @@ -802,7 +816,7 @@ static int virtnet_xdp_xmit(struct net_device *dev,
> > return ret;
> > }
> >
> > -static void put_xdp_frags(struct xdp_buff *xdp)
> > +static void put_xdp_frags(struct xdp_buff *xdp, struct receive_queue *rq)
> > {
>
> rq could be fetched from xdp_rxq_info?

Yeah, it has the queue_index there.
>
> > struct skb_shared_info *shinfo;
> > struct page *xdp_page;
> > @@ -812,7 +826,7 @@ static void put_xdp_frags(struct xdp_buff *xdp)
> > shinfo = xdp_get_shared_info_from_buff(xdp);
> > for (i = 0; i < shinfo->nr_frags; i++) {
> > xdp_page = skb_frag_page(&shinfo->frags[i]);
> > - put_page(xdp_page);
> > + virtnet_put_page(rq, xdp_page);
> > }
> > }
> > }
> > @@ -903,7 +917,11 @@ static struct page *xdp_linearize_page(struct receive_queue *rq,
> > if (page_off + *len + tailroom > PAGE_SIZE)
> > return NULL;
> >
> > - page = alloc_page(GFP_ATOMIC);
> > + if (rq->page_pool)
> > + page = page_pool_dev_alloc_pages(rq->page_pool);
> > + else
> > + page = alloc_page(GFP_ATOMIC);
> > +
> > if (!page)
> > return NULL;
> >
> > @@ -926,21 +944,24 @@ static struct page *xdp_linearize_page(struct receive_queue *rq,
> > * is sending packet larger than the MTU.
> > */
> > if ((page_off + buflen + tailroom) > PAGE_SIZE) {
> > - put_page(p);
> > + virtnet_put_page(rq, p);
> > goto err_buf;
> > }
> >
> > memcpy(page_address(page) + page_off,
> > page_address(p) + off, buflen);
> > page_off += buflen;
> > - put_page(p);
> > + virtnet_put_page(rq, p);
> > }
> >
> > /* Headroom does not contribute to packet length */
> > *len = page_off - VIRTIO_XDP_HEADROOM;
> > return page;
> > err_buf:
> > - __free_pages(page, 0);
> > + if (rq->page_pool)
> > + page_pool_put_full_page(rq->page_pool, page, true);
> > + else
> > + __free_pages(page, 0);
> > return NULL;
> > }
> >
> > @@ -1144,7 +1165,7 @@ static void mergeable_buf_free(struct receive_queue *rq, int num_buf,
> > }
> > stats->bytes += len;
> > page = virt_to_head_page(buf);
> > - put_page(page);
> > + virtnet_put_page(rq, page);
> > }
> > }
> >
> > @@ -1264,7 +1285,7 @@ static int virtnet_build_xdp_buff_mrg(struct net_device *dev,
> > cur_frag_size = truesize;
> > xdp_frags_truesz += cur_frag_size;
> > if (unlikely(len > truesize - room || cur_frag_size > PAGE_SIZE)) {
> > - put_page(page);
> > + virtnet_put_page(rq, page);
> > pr_debug("%s: rx error: len %u exceeds truesize %lu\n",
> > dev->name, len, (unsigned long)(truesize - room));
> > dev->stats.rx_length_errors++;
> > @@ -1283,7 +1304,7 @@ static int virtnet_build_xdp_buff_mrg(struct net_device *dev,
> > return 0;
> >
> > err:
> > - put_xdp_frags(xdp);
> > + put_xdp_frags(xdp, rq);
> > return -EINVAL;
> > }
> >
> > @@ -1344,7 +1365,10 @@ static void *mergeable_xdp_get_buf(struct virtnet_info *vi,
> > if (*len + xdp_room > PAGE_SIZE)
> > return NULL;
> >
> > - xdp_page = alloc_page(GFP_ATOMIC);
> > + if (rq->page_pool)
> > + xdp_page = page_pool_dev_alloc_pages(rq->page_pool);
> > + else
> > + xdp_page = alloc_page(GFP_ATOMIC);
> > if (!xdp_page)
> > return NULL;
> >
> > @@ -1354,7 +1378,7 @@ static void *mergeable_xdp_get_buf(struct virtnet_info *vi,
> >
> > *frame_sz = PAGE_SIZE;
> >
> > - put_page(*page);
> > + virtnet_put_page(rq, *page);
> >
> > *page = xdp_page;
> >
> > @@ -1400,6 +1424,8 @@ static struct sk_buff *receive_mergeable_xdp(struct net_device *dev,
> > head_skb = build_skb_from_xdp_buff(dev, vi, &xdp, xdp_frags_truesz);
> > if (unlikely(!head_skb))
> > break;
> > + if (rq->page_pool)
> > + skb_mark_for_recycle(head_skb);
> > return head_skb;
> >
> > case XDP_TX:
> > @@ -1410,10 +1436,10 @@ static struct sk_buff *receive_mergeable_xdp(struct net_device *dev,
> > break;
> > }
> >
> > - put_xdp_frags(&xdp);
> > + put_xdp_frags(&xdp, rq);
> >
> > err_xdp:
> > - put_page(page);
> > + virtnet_put_page(rq, page);
> > mergeable_buf_free(rq, num_buf, dev, stats);
> >
> > stats->xdp_drops++;
> > @@ -1467,6 +1493,9 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
> > head_skb = page_to_skb(vi, rq, page, offset, len, truesize, headroom);
> > curr_skb = head_skb;
> >
> > + if (rq->page_pool)
> > + skb_mark_for_recycle(curr_skb);
> > +
> > if (unlikely(!curr_skb))
> > goto err_skb;
> > while (--num_buf) {
> > @@ -1509,6 +1538,8 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
> > curr_skb = nskb;
> > head_skb->truesize += nskb->truesize;
> > num_skb_frags = 0;
> > + if (rq->page_pool)
> > + skb_mark_for_recycle(curr_skb);
> > }
> > if (curr_skb != head_skb) {
> > head_skb->data_len += len;
> > @@ -1517,7 +1548,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
> > }
> > offset = buf - page_address(page);
> > if (skb_can_coalesce(curr_skb, num_skb_frags, page, offset)) {
> > - put_page(page);
> > + virtnet_put_page(rq, page);
>
> I wonder why not we can't do this during buffer allocation like other drivers?
>

Sorry, I don't quite understand the point here. Would you please
elaborate a bit more?
> > skb_coalesce_rx_frag(curr_skb, num_skb_frags - 1,
> > len, truesize);
> > } else {
> > @@ -1530,7 +1561,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
> > return head_skb;
> >
> > err_skb:
> > - put_page(page);
> > + virtnet_put_page(rq, page);
> > mergeable_buf_free(rq, num_buf, dev, stats);
> >
> > err_buf:
> > @@ -1737,31 +1768,40 @@ static int add_recvbuf_mergeable(struct virtnet_info *vi,
> > * disabled GSO for XDP, it won't be a big issue.
> > */
> > len = get_mergeable_buf_len(rq, &rq->mrg_avg_pkt_len, room);
> > - if (unlikely(!skb_page_frag_refill(len + room, alloc_frag, gfp)))
> > - return -ENOMEM;
> > + if (rq->page_pool) {
> > + struct page *page;
> >
> > - buf = (char *)page_address(alloc_frag->page) + alloc_frag->offset;
> > - buf += headroom; /* advance address leaving hole at front of pkt */
> > - get_page(alloc_frag->page);
> > - alloc_frag->offset += len + room;
> > - hole = alloc_frag->size - alloc_frag->offset;
> > - if (hole < len + room) {
> > - /* To avoid internal fragmentation, if there is very likely not
> > - * enough space for another buffer, add the remaining space to
> > - * the current buffer.
> > - * XDP core assumes that frame_size of xdp_buff and the length
> > - * of the frag are PAGE_SIZE, so we disable the hole mechanism.
> > - */
> > - if (!headroom)
> > - len += hole;
> > - alloc_frag->offset += hole;
> > - }
> > + page = page_pool_dev_alloc_pages(rq->page_pool);
> > + if (unlikely(!page))
> > + return -ENOMEM;
> > + buf = (char *)page_address(page);
> > + buf += headroom; /* advance address leaving hole at front of pkt */
> > + } else {
> > + if (unlikely(!skb_page_frag_refill(len + room, alloc_frag, gfp)))
>
> Why not simply use a helper like virtnet_page_frag_refill() and add
> the page_pool allocation logic there? It helps to reduce the
> changeset.
>

Sure. Will do that on v2.
> > + return -ENOMEM;
> >
> > + buf = (char *)page_address(alloc_frag->page) + alloc_frag->offset;
> > + buf += headroom; /* advance address leaving hole at front of pkt */
> > + get_page(alloc_frag->page);
> > + alloc_frag->offset += len + room;
> > + hole = alloc_frag->size - alloc_frag->offset;
> > + if (hole < len + room) {
> > + /* To avoid internal fragmentation, if there is very likely not
> > + * enough space for another buffer, add the remaining space to
> > + * the current buffer.
> > + * XDP core assumes that frame_size of xdp_buff and the length
> > + * of the frag are PAGE_SIZE, so we disable the hole mechanism.
> > + */
> > + if (!headroom)
> > + len += hole;
> > + alloc_frag->offset += hole;
> > + }
> > + }
> > sg_init_one(rq->sg, buf, len);
> > ctx = mergeable_len_to_ctx(len + room, headroom);
> > err = virtqueue_add_inbuf_ctx(rq->vq, rq->sg, 1, buf, ctx, gfp);
> > if (err < 0)
> > - put_page(virt_to_head_page(buf));
> > + virtnet_put_page(rq, virt_to_head_page(buf));
> >
> > return err;
> > }
> > @@ -1994,8 +2034,15 @@ static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index)
> > if (err < 0)
> > return err;
> >
> > - err = xdp_rxq_info_reg_mem_model(&vi->rq[qp_index].xdp_rxq,
> > - MEM_TYPE_PAGE_SHARED, NULL);
> > + if (vi->rq[qp_index].page_pool)
> > + err = xdp_rxq_info_reg_mem_model(&vi->rq[qp_index].xdp_rxq,
> > + MEM_TYPE_PAGE_POOL,
> > + vi->rq[qp_index].page_pool);
> > + else
> > + err = xdp_rxq_info_reg_mem_model(&vi->rq[qp_index].xdp_rxq,
> > + MEM_TYPE_PAGE_SHARED,
> > + NULL);
> > +
> > if (err < 0)
> > goto err_xdp_reg_mem_model;
> >
> > @@ -2951,6 +2998,7 @@ static void virtnet_get_strings(struct net_device *dev, u32 stringset, u8 *data)
> > ethtool_sprintf(&p, "tx_queue_%u_%s", i,
> > virtnet_sq_stats_desc[j].desc);
> > }
> > + page_pool_ethtool_stats_get_strings(p);
> > break;
> > }
> > }
> > @@ -2962,12 +3010,30 @@ static int virtnet_get_sset_count(struct net_device *dev, int sset)
> > switch (sset) {
> > case ETH_SS_STATS:
> > return vi->curr_queue_pairs * (VIRTNET_RQ_STATS_LEN +
> > - VIRTNET_SQ_STATS_LEN);
> > + VIRTNET_SQ_STATS_LEN +
> > + (page_pool_enabled && vi->mergeable_rx_bufs ?
> > + page_pool_ethtool_stats_get_count() : 0));
> > default:
> > return -EOPNOTSUPP;
> > }
> > }
> >
> > +static void virtnet_get_page_pool_stats(struct net_device *dev, u64 *data)
> > +{
> > +#ifdef CONFIG_PAGE_POOL_STATS
> > + struct virtnet_info *vi = netdev_priv(dev);
> > + struct page_pool_stats pp_stats = {};
> > + int i;
> > +
> > + for (i = 0; i < vi->curr_queue_pairs; i++) {
> > + if (!vi->rq[i].page_pool)
> > + continue;
> > + page_pool_get_stats(vi->rq[i].page_pool, &pp_stats);
> > + }
> > + page_pool_ethtool_stats_get(data, &pp_stats);
> > +#endif /* CONFIG_PAGE_POOL_STATS */
> > +}
> > +
> > static void virtnet_get_ethtool_stats(struct net_device *dev,
> > struct ethtool_stats *stats, u64 *data)
> > {
> > @@ -3003,6 +3069,8 @@ static void virtnet_get_ethtool_stats(struct net_device *dev,
> > } while (u64_stats_fetch_retry(&sq->stats.syncp, start));
> > idx += VIRTNET_SQ_STATS_LEN;
> > }
> > +
> > + virtnet_get_page_pool_stats(dev, &data[idx]);
> > }
> >
> > static void virtnet_get_channels(struct net_device *dev,
> > @@ -3623,6 +3691,8 @@ static void virtnet_free_queues(struct virtnet_info *vi)
> > for (i = 0; i < vi->max_queue_pairs; i++) {
> > __netif_napi_del(&vi->rq[i].napi);
> > __netif_napi_del(&vi->sq[i].napi);
> > + if (vi->rq[i].page_pool)
> > + page_pool_destroy(vi->rq[i].page_pool);
> > }
> >
> > /* We called __netif_napi_del(),
> > @@ -3679,12 +3749,19 @@ static void virtnet_rq_free_unused_buf(struct virtqueue *vq, void *buf)
> > struct virtnet_info *vi = vq->vdev->priv;
> > int i = vq2rxq(vq);
> >
> > - if (vi->mergeable_rx_bufs)
> > - put_page(virt_to_head_page(buf));
> > - else if (vi->big_packets)
> > + if (vi->mergeable_rx_bufs) {
> > + if (vi->rq[i].page_pool) {
> > + page_pool_put_full_page(vi->rq[i].page_pool,
> > + virt_to_head_page(buf),
> > + true);
> > + } else {
> > + put_page(virt_to_head_page(buf));
> > + }
> > + } else if (vi->big_packets) {
> > give_pages(&vi->rq[i], buf);
>
> Any reason only mergeable were modified but not for small and big?
>
> Thanks
>

Big mode uses the page chain to recycle pages, thus the using of
"private" of the buffer page. I will take further look into that to
see if it is better to use page pool in these cases. Thanks!

> > - else
> > + } else {
> > put_page(virt_to_head_page(buf));
> > + }
> > }
> >
> > static void free_unused_bufs(struct virtnet_info *vi)
> > @@ -3718,6 +3795,26 @@ static void virtnet_del_vqs(struct virtnet_info *vi)
> > virtnet_free_queues(vi);
> > }
> >
> > +static void virtnet_alloc_page_pool(struct receive_queue *rq)
> > +{
> > + struct virtio_device *vdev = rq->vq->vdev;
> > +
> > + struct page_pool_params pp_params = {
> > + .order = 0,
> > + .pool_size = rq->vq->num_max,
> > + .nid = dev_to_node(vdev->dev.parent),
> > + .dev = vdev->dev.parent,
> > + .offset = 0,
> > + };
> > +
> > + rq->page_pool = page_pool_create(&pp_params);
> > + if (IS_ERR(rq->page_pool)) {
> > + dev_warn(&vdev->dev, "page pool creation failed: %ld\n",
> > + PTR_ERR(rq->page_pool));
> > + rq->page_pool = NULL;
> > + }
> > +}
> > +
> > /* How large should a single buffer be so a queue full of these can fit at
> > * least one full packet?
> > * Logic below assumes the mergeable buffer header is used.
> > @@ -3801,6 +3898,13 @@ static int virtnet_find_vqs(struct virtnet_info *vi)
> > vi->rq[i].vq = vqs[rxq2vq(i)];
> > vi->rq[i].min_buf_len = mergeable_min_buf_len(vi, vi->rq[i].vq);
> > vi->sq[i].vq = vqs[txq2vq(i)];
> > +
> > + if (page_pool_enabled && vi->mergeable_rx_bufs)
> > + virtnet_alloc_page_pool(&vi->rq[i]);
> > + else
> > + dev_warn(&vi->vdev->dev,
> > + "page pool only support mergeable mode\n");
> > +
> > }
> >
> > /* run here: ret == 0. */
> > --
> > 2.31.1
> >
>

Next message: Arnaldo Carvalho de Melo: "Re: [PATCH v1] perf test python: Put perf python at start of sys.path"
Previous message: Liang Chen: "Re: [PATCH net-next 1/5] virtio_net: Fix an unsafe reference to the page chain"
In reply to: Jason Wang: "Re: [PATCH net-next 2/5] virtio_net: Add page_pool support to improve performance"
Next in thread: Michael S. Tsirkin: "Re: [PATCH net-next 2/5] virtio_net: Add page_pool support to improve performance"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]