From: "Yu, Jin" <jin.yu@intel.com>
To: "Bie, Tiwei" <tiwei.bie@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
"Liu, Changpeng" <changpeng.liu@intel.com>,
"maxime.coquelin@redhat.com" <maxime.coquelin@redhat.com>,
"Wang, Zhihong" <zhihong.wang@intel.com>,
Lin Li <lilin24@baidu.com>, Xun Ni <nixun@baidu.com>,
Yu Zhang <zhangyu31@baidu.com>
Subject: Re: [dpdk-dev] [PATCH v7 05/10] vhost: checkout the resubmit inflight information
Date: Thu, 26 Sep 2019 15:52:20 +0000 [thread overview]
Message-ID: <B9FBC361811A3D4DBB02350807E29F7B0B95B705@SHSMSX101.ccr.corp.intel.com> (raw)
In-Reply-To: <20190926080440.GA17894@___>
> -----Original Message-----
> From: Bie, Tiwei
> Sent: Thursday, September 26, 2019 4:05 PM
> To: Yu, Jin <jin.yu@intel.com>
> Cc: dev@dpdk.org; Liu, Changpeng <changpeng.liu@intel.com>;
> maxime.coquelin@redhat.com; Wang, Zhihong <zhihong.wang@intel.com>; Lin
> Li <lilin24@baidu.com>; Xun Ni <nixun@baidu.com>; Yu Zhang
> <zhangyu31@baidu.com>
> Subject: Re: [PATCH v7 05/10] vhost: checkout the resubmit inflight information
>
> On Fri, Sep 20, 2019 at 08:00:57PM +0800, Jin Yu wrote:
> > @@ -1458,6 +1472,188 @@ static int vhost_user_set_vring_err(struct
> virtio_net **pdev __rte_unused,
> > return RTE_VHOST_MSG_RESULT_OK;
> > }
> >
> > +static int
> > +resubmit_desc_compare(const void *a, const void *b) {
> > + const struct rte_vhost_resubmit_desc *desc0 =
> > + (const struct rte_vhost_resubmit_desc *)a;
> > + const struct rte_vhost_resubmit_desc *desc1 =
> > + (const struct rte_vhost_resubmit_desc *)b;
> > +
> > + if (desc1->counter > desc0->counter &&
> > + (desc1->counter - desc0->counter) < VIRTQUEUE_MAX_SIZE * 2)
>
> Why is this hardcoded to 1024 * 2 here?
I will check this. Thanks.
>
> > + return 1;
> > +
> > + return -1;
> > +}
> > +
> > +static int
> > +vhost_check_queue_inflights_split(struct virtio_net *dev,
> > + struct vhost_virtqueue *vq)
> > +{
> > + uint16_t i = 0;
> > + uint16_t resubmit_num = 0, last_io, num;
> > + struct vring_used *used = vq->used;
> > + struct rte_vhost_resubmit_info *resubmit = NULL;
> > + struct rte_vhost_inflight_info_split *inflight_split;
> > +
> > + if (!(dev->protocol_features &
> > + (1ULL << VHOST_USER_PROTOCOL_F_INFLIGHT_SHMFD)))
> > + return RTE_VHOST_MSG_RESULT_OK;
> > +
> > + if ((!vq->inflight_split))
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > +
> > + if (!vq->inflight_split->version) {
> > + vq->inflight_split->version = INFLIGHT_VERSION;
> > + return RTE_VHOST_MSG_RESULT_OK;
> > + }
> > +
> > + inflight_split = vq->inflight_split;
> > + vq->resubmit_inflight = NULL;
>
> This is invoked in vhost message handler, you need to check whether it has been
> allocated first.
Got it, Thanks.
>
>
> > + vq->global_counter = 0;
> > + last_io = inflight_split->last_inflight_io;
> > +
> > + if (inflight_split->used_idx != used->idx) {
> > + inflight_split->desc[last_io].inflight = 0;
> > + rte_compiler_barrier();
> > + inflight_split->used_idx = used->idx;
> > + }
> > +
> > + for (i = 0; i < inflight_split->desc_num; i++) {
> > + if (inflight_split->desc[i].inflight == 1)
> > + resubmit_num++;
> > + }
> > +
> > + vq->last_avail_idx += resubmit_num;
> > +
> > + if (resubmit_num) {
> > + resubmit = calloc(1, sizeof(struct rte_vhost_resubmit_info));
> > + if (!resubmit) {
> > + RTE_LOG(ERR, VHOST_CONFIG,
> > + "Failed to allocate memory for resubmit info.\n");
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > + }
> > +
> > + resubmit->resubmit_list = calloc(resubmit_num,
> > + sizeof(struct rte_vhost_resubmit_desc));
> > + if (!resubmit->resubmit_list) {
> > + RTE_LOG(ERR, VHOST_CONFIG,
> > + "Failed to allocate memory for inflight desc.\n");
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > + }
> > +
> > + num = 0;
> > + for (i = 0; i < vq->inflight_split->desc_num; i++) {
> > + if (vq->inflight_split->desc[i].inflight == 1) {
> > + resubmit->resubmit_list[num].index = i;
> > + resubmit->resubmit_list[num].counter =
> > + inflight_split->desc[i].counter;
> > + num++;
> > + }
> > + }
> > + resubmit->resubmit_num = num;
> > +
> > + if (resubmit->resubmit_num > 1)
> > + qsort(resubmit->resubmit_list, resubmit-
> >resubmit_num,
> > + sizeof(struct rte_vhost_resubmit_desc),
> > + resubmit_desc_compare);
> > +
> > + vq->global_counter = resubmit->resubmit_list[0].counter + 1;
> > + vq->resubmit_inflight = resubmit;
> > + }
> > +
> > + return RTE_VHOST_MSG_RESULT_OK;
> > +}
> > +
> > +static int
> > +vhost_check_queue_inflights_packed(struct virtio_net *dev,
> > + struct vhost_virtqueue *vq)
> > +{
> > + uint16_t i = 0;
> > + uint16_t resubmit_num = 0, old_used_idx, num;
> > + struct rte_vhost_resubmit_info *resubmit = NULL;
> > + struct rte_vhost_inflight_info_packed *inflight_packed;
> > +
> > + if (!(dev->protocol_features &
> > + (1ULL << VHOST_USER_PROTOCOL_F_INFLIGHT_SHMFD)))
> > + return RTE_VHOST_MSG_RESULT_OK;
> > +
> > + if ((!vq->inflight_packed))
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > +
> > + if (!vq->inflight_packed->version) {
> > + vq->inflight_packed->version = INFLIGHT_VERSION;
> > + return RTE_VHOST_MSG_RESULT_OK;
> > + }
> > +
> > + inflight_packed = vq->inflight_packed;
> > + vq->resubmit_inflight = NULL;
>
> Ditto.
>
>
> > + vq->global_counter = 0;
> > + old_used_idx = inflight_packed->old_used_idx;
> > +
> > + if (inflight_packed->used_idx != old_used_idx) {
> > + if (inflight_packed->desc[old_used_idx].inflight == 0) {
> > + inflight_packed->old_used_idx =
> > + inflight_packed->used_idx;
> > + inflight_packed->old_used_wrap_counter =
> > + inflight_packed->used_wrap_counter;
> > + inflight_packed->old_free_head =
> > + inflight_packed->free_head;
> > + } else {
> > + inflight_packed->used_idx =
> > + inflight_packed->old_used_idx;
> > + inflight_packed->used_wrap_counter =
> > + inflight_packed->old_used_wrap_counter;
> > + inflight_packed->free_head =
> > + inflight_packed->old_free_head;
> > + }
> > + }
> > +
> > + for (i = 0; i < inflight_packed->desc_num; i++) {
> > + if (inflight_packed->desc[i].inflight == 1)
> > + resubmit_num++;
> > + }
> > +
> > + if (resubmit_num) {
> > + resubmit = calloc(1, sizeof(struct rte_vhost_resubmit_info));
> > + if (resubmit == NULL) {
> > + RTE_LOG(ERR, VHOST_CONFIG,
> > + "Failed to allocate memory for resubmit info.\n");
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > + }
> > +
> > + resubmit->resubmit_list = calloc(resubmit_num,
> > + sizeof(struct rte_vhost_resubmit_desc));
> > + if (resubmit->resubmit_list == NULL) {
> > + RTE_LOG(ERR, VHOST_CONFIG,
> > + "Failed to allocate memory for resubmit desc.\n");
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > + }
> > +
> > + num = 0;
> > + for (i = 0; i < inflight_packed->desc_num; i++) {
> > + if (vq->inflight_packed->desc[i].inflight == 1) {
> > + resubmit->resubmit_list[num].index = i;
> > + resubmit->resubmit_list[num].counter =
> > + inflight_packed->desc[i].counter;
> > + num++;
> > + }
> > + }
> > + resubmit->resubmit_num = num;
> > +
> > + if (resubmit->resubmit_num > 1)
> > + qsort(resubmit->resubmit_list, resubmit-
> >resubmit_num,
> > + sizeof(struct rte_vhost_resubmit_desc),
> > + resubmit_desc_compare);
> > +
> > + vq->global_counter = resubmit->resubmit_list[0].counter + 1;
> > + vq->resubmit_inflight = resubmit;
> > + }
> > +
> > + return RTE_VHOST_MSG_RESULT_OK;
> > +}
> > +
> > static int
> > vhost_user_set_vring_kick(struct virtio_net **pdev, struct VhostUserMsg
> *msg,
> > int main_fd __rte_unused)
> > @@ -1499,6 +1695,20 @@ vhost_user_set_vring_kick(struct virtio_net
> **pdev, struct VhostUserMsg *msg,
> > close(vq->kickfd);
> > vq->kickfd = file.fd;
> >
> > + if (vq_is_packed(dev)) {
> > + if (vhost_check_queue_inflights_packed(dev, vq)) {
> > + RTE_LOG(ERR, VHOST_CONFIG,
> > + "Failed to inflights for vq: %d\n", file.index);
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > + }
> > + } else {
> > + if (vhost_check_queue_inflights_split(dev, vq)) {
> > + RTE_LOG(ERR, VHOST_CONFIG,
> > + "Failed to inflights for vq: %d\n", file.index);
> > + return RTE_VHOST_MSG_RESULT_ERR;
> > + }
> > + }
> > +
> > return RTE_VHOST_MSG_RESULT_OK;
> > }
> >
> > --
> > 2.17.2
> >
next prev parent reply other threads:[~2019-09-26 15:52 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20190917145234.16951>
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 00/10] vhost: support inflight share memory protocol feature Jin Yu
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 01/10] vhost: add the inflight description Jin Yu
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 02/10] vhost: add packed ring Jin Yu
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 03/10] vhost: add the inflight structure Jin Yu
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 04/10] vhost: add two new messages to support a shared buffer Jin Yu
2019-09-26 7:39 ` Tiwei Bie
2019-09-26 15:06 ` Yu, Jin
2019-09-27 2:12 ` Tiwei Bie
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 05/10] vhost: checkout the resubmit inflight information Jin Yu
2019-09-26 8:04 ` Tiwei Bie
2019-09-26 15:52 ` Yu, Jin [this message]
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 06/10] vhost: add the APIs to operate inflight ring Jin Yu
2019-09-20 12:00 ` [dpdk-dev] [PATCH v7 07/10] vhost: add APIs for user getting " Jin Yu
2019-09-20 12:01 ` [dpdk-dev] [PATCH v7 08/10] vhost: add vring functions packed ring support Jin Yu
2019-09-20 12:01 ` [dpdk-dev] [PATCH v7 09/10] vhost: add an API for judging vq format Jin Yu
2019-09-20 12:01 ` [dpdk-dev] [PATCH v7 10/10] vhost: add vhost-user-blk example which support inflight Jin Yu
2019-09-25 14:45 ` Tiwei Bie
2019-09-26 14:29 ` Yu, Jin
2019-09-26 14:40 ` Tiwei Bie
2019-09-25 14:25 ` [dpdk-dev] [PATCH v7 00/10] vhost: support inflight share memory protocol feature Tiwei Bie
2019-09-26 14:00 ` Yu, Jin
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=B9FBC361811A3D4DBB02350807E29F7B0B95B705@SHSMSX101.ccr.corp.intel.com \
--to=jin.yu@intel.com \
--cc=changpeng.liu@intel.com \
--cc=dev@dpdk.org \
--cc=lilin24@baidu.com \
--cc=maxime.coquelin@redhat.com \
--cc=nixun@baidu.com \
--cc=tiwei.bie@intel.com \
--cc=zhangyu31@baidu.com \
--cc=zhihong.wang@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).