From: "Yao, Lei A" <lei.a.yao@intel.com>
To: "Chen, Junjie J" <junjie.j.chen@intel.com>,
"yliu@fridaylinux.org" <yliu@fridaylinux.org>,
"maxime.coquelin@redhat.com" <maxime.coquelin@redhat.com>,
"Wang, Xiao W" <xiao.w.wang@intel.com>,
"Bie, Tiwei" <tiwei.bie@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v7] vhost: support virtqueue interrupt/notification suppression
Date: Tue, 9 Jan 2018 07:34:17 +0000 [thread overview]
Message-ID: <2DBBFF226F7CF64BAFCA79B681719D953A3220A3@shsmsx102.ccr.corp.intel.com> (raw)
In-Reply-To: <1515495828-190409-1-git-send-email-junjie.j.chen@intel.com>
> -----Original Message-----
> From: Chen, Junjie J
> Sent: Tuesday, January 9, 2018 7:04 PM
> To: yliu@fridaylinux.org; maxime.coquelin@redhat.com; Wang, Xiao W
> <xiao.w.wang@intel.com>; Bie, Tiwei <tiwei.bie@intel.com>; Yao, Lei A
> <lei.a.yao@intel.com>
> Cc: dev@dpdk.org; Chen, Junjie J <junjie.j.chen@intel.com>
> Subject: [PATCH v7] vhost: support virtqueue interrupt/notification
> suppression
>
> The driver can suppress interrupt when VIRTIO_F_EVENT_IDX feature bit is
> negotiated. The driver set vring flags to 0, and MAY use used_event in
> available ring to advise device interrupt util reach an index specified
> by used_event. The device ignore the lower bit of vring flags, and send
> an interrupt when index reach used_event.
>
> The device can suppress notification in a manner analogous to the ways
> driver suppress interrupt. The device manipulates flags or avail_event in
> the used ring in the same way the driver manipulates flags or used_event in
> available ring.
>
> Signed-off-by: Junjie Chen <junjie.j.chen@intel.com>
Tested-by: Lei Yao <lei.a.yao@intel.com>
VM2VM Iperf test has been executed with virtio-net driver after apply this patch.
No performance drop.
CPU: Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
Host OS: Ubuntu 16.04
Guest OS: Ubuntu 16.04
Kernel: 4.4.0
> ---
> v7:
> Add vhost_need_event definition and update code for next virtio.
>
> v6:
> Use volatile qualifier to access avail event idx.
>
> v5:
> Remove updating avail event index in backend.
>
> v2-v4:
> Use definition of VIRTIO_F_EVENT_IDX from kernel.
>
> lib/librte_vhost/vhost.c | 2 +-
> lib/librte_vhost/vhost.h | 44
> ++++++++++++++++++++++++++++++++++++++-----
> lib/librte_vhost/virtio_net.c | 6 +++---
> 3 files changed, 43 insertions(+), 9 deletions(-)
>
> diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
> index 6f7ef7f..400caa0 100644
> --- a/lib/librte_vhost/vhost.c
> +++ b/lib/librte_vhost/vhost.c
> @@ -538,7 +538,7 @@ rte_vhost_vring_call(int vid, uint16_t vring_idx)
> if (!vq)
> return -1;
>
> - vhost_vring_call(vq);
> + vhost_vring_call(dev, vq);
> return 0;
> }
>
> diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
> index 1d9366e..ec79991 100644
> --- a/lib/librte_vhost/vhost.h
> +++ b/lib/librte_vhost/vhost.h
> @@ -103,6 +103,8 @@ struct vhost_virtqueue {
>
> uint16_t last_avail_idx;
> uint16_t last_used_idx;
> + /* Last used index we notify to front end. */
> + uint16_t signalled_used;
> #define VIRTIO_INVALID_EVENTFD (-1)
> #define VIRTIO_UNINITIALIZED_EVENTFD (-2)
>
> @@ -214,6 +216,7 @@ struct vhost_msg {
> (1ULL << VIRTIO_NET_F_GUEST_TSO6) | \
> (1ULL << VIRTIO_NET_F_GUEST_UFO) | \
> (1ULL << VIRTIO_RING_F_INDIRECT_DESC) |
> \
> + (1ULL << VIRTIO_RING_F_EVENT_IDX) | \
> (1ULL << VIRTIO_NET_F_MTU) | \
> (1ULL << VIRTIO_F_IOMMU_PLATFORM))
>
> @@ -399,16 +402,47 @@ vhost_iova_to_vva(struct virtio_net *dev, struct
> vhost_virtqueue *vq,
> return __vhost_iova_to_vva(dev, vq, iova, size, perm);
> }
>
> +#define vhost_used_event(vr) \
> + (*(volatile uint16_t*)&(vr)->avail->ring[(vr)->size])
> +
> +/*
> + * The following is used with VIRTIO_RING_F_EVENT_IDX.
> + * Assuming a given event_idx value from the other size, if we have
> + * just incremented index from old to new_idx, should we trigger an
> + * event?
> + */
> +static __rte_always_inline int
> +vhost_need_event(uint16_t event_idx, uint16_t new_idx, uint16_t old)
> +{
> + return (uint16_t)(new_idx - event_idx - 1) < (uint16_t)(new_idx -
> old);
> +}
> +
> static __rte_always_inline void
> -vhost_vring_call(struct vhost_virtqueue *vq)
> +vhost_vring_call(struct virtio_net *dev, struct vhost_virtqueue *vq)
> {
> /* Flush used->idx update before we read avail->flags. */
> 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);
> + /* Don't kick guest if we don't reach index specified by guest. */
> + if (dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX)) {
> + uint16_t old = vq->signalled_used;
> + uint16_t new = vq->last_used_idx;
> +
> + LOG_DEBUG(VHOST_DATA, "%s: used_event_idx=%d,
> old=%d, new=%d\n",
> + __func__,
> + vhost_used_event(vq),
> + old, new);
> + if (vhost_need_event(vhost_used_event(vq), new, old)
> + && (vq->callfd >= 0)) {
> + vq->signalled_used = vq->last_used_idx;
> + eventfd_write(vq->callfd, (eventfd_t) 1);
> + }
> + } else {
> + /* Kick the guest if necessary. */
> + if (!(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
> + && (vq->callfd >= 0))
> + eventfd_write(vq->callfd, (eventfd_t)1);
> + }
> }
>
> #endif /* _VHOST_NET_CDEV_H_ */
> diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
> index eb496e1..d0f9221 100644
> --- a/lib/librte_vhost/virtio_net.c
> +++ b/lib/librte_vhost/virtio_net.c
> @@ -413,7 +413,7 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t
> queue_id,
> offsetof(struct vring_used, idx),
> sizeof(vq->used->idx));
>
> - vhost_vring_call(vq);
> + vhost_vring_call(dev, vq);
> out:
> if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
> vhost_user_iotlb_rd_unlock(vq);
> @@ -700,7 +700,7 @@ virtio_dev_merge_rx(struct virtio_net *dev, uint16_t
> queue_id,
>
> if (likely(vq->shadow_used_idx)) {
> flush_shadow_used_ring(dev, vq);
> - vhost_vring_call(vq);
> + vhost_vring_call(dev, vq);
> }
>
> out:
> @@ -1104,7 +1104,7 @@ update_used_idx(struct virtio_net *dev, struct
> vhost_virtqueue *vq,
> vq->used->idx += count;
> vhost_log_used_vring(dev, vq, offsetof(struct vring_used, idx),
> sizeof(vq->used->idx));
> - vhost_vring_call(vq);
> + vhost_vring_call(dev, vq);
> }
>
> static __rte_always_inline struct zcopy_mbuf *
> --
> 2.0.1
next prev parent reply other threads:[~2018-01-09 7:34 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-28 11:02 [dpdk-dev] [PATCH v4] " junjie.j.chen
2017-12-23 16:55 ` [dpdk-dev] [PATCH v5] " Junjie Chen
2017-12-26 8:03 ` Wang, Xiao W
2017-12-26 17:43 ` [dpdk-dev] [PATCH v6] " Junjie Chen
2018-01-05 5:25 ` Wang, Xiao W
2018-01-08 14:06 ` Yuanhan Liu
2018-01-09 2:12 ` Chen, Junjie J
2018-01-09 2:29 ` Yuanhan Liu
2018-01-09 11:03 ` [dpdk-dev] [PATCH v7] " Junjie Chen
2018-01-09 7:34 ` Yao, Lei A [this message]
2018-01-09 12:58 ` Yuanhan Liu
2018-01-09 8:56 ` Maxime Coquelin
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=2DBBFF226F7CF64BAFCA79B681719D953A3220A3@shsmsx102.ccr.corp.intel.com \
--to=lei.a.yao@intel.com \
--cc=dev@dpdk.org \
--cc=junjie.j.chen@intel.com \
--cc=maxime.coquelin@redhat.com \
--cc=tiwei.bie@intel.com \
--cc=xiao.w.wang@intel.com \
--cc=yliu@fridaylinux.org \
/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).