From: Maxime Coquelin <maxime.coquelin@redhat.com>
To: Jens Freimann <jfreimann@redhat.com>, Tiwei Bie <tiwei.bie@intel.com>
Cc: dev@dpdk.org, zhihong.wang@intel.com
Subject: Re: [dpdk-dev] [PATCH v9 1/8] net/virtio: vring init for packed queues
Date: Thu, 25 Oct 2018 15:51:19 +0200 [thread overview]
Message-ID: <931e66a9-5f22-e241-0cbe-4640698543ec@redhat.com> (raw)
In-Reply-To: <20181025134737.vzstdn5td7b55nej@jenstp.localdomain>
On 10/25/18 3:48 PM, Jens Freimann wrote:
> On Thu, Oct 25, 2018 at 05:21:15PM +0800, Tiwei Bie wrote:
>> On Wed, Oct 24, 2018 at 04:32:29PM +0200, Jens Freimann wrote:
>>> Add and initialize descriptor data structures.
>>>
>>> Signed-off-by: Jens Freimann <jfreimann@redhat.com>
>>> ---
>>> drivers/net/virtio/virtio_ethdev.c | 18 ++++----
>>> drivers/net/virtio/virtio_pci.h | 7 +++
>>> drivers/net/virtio/virtio_ring.h | 74 ++++++++++++++++++++++++++----
>>> drivers/net/virtio/virtqueue.h | 15 +++++-
>>> 4 files changed, 95 insertions(+), 19 deletions(-)
>>>
>>> diff --git a/drivers/net/virtio/virtio_ethdev.c
>>> b/drivers/net/virtio/virtio_ethdev.c
>>> index 730c41707..6130aef16 100644
>>> --- a/drivers/net/virtio/virtio_ethdev.c
>>> +++ b/drivers/net/virtio/virtio_ethdev.c
>>> @@ -299,19 +299,21 @@ virtio_init_vring(struct virtqueue *vq)
>>>
>>> PMD_INIT_FUNC_TRACE();
>>>
>>> - /*
>>> - * Reinitialise since virtio port might have been stopped and
>>> restarted
>>> - */
>>
>> Why above comments are removed?
>
> by mistake, I'll put them back in next version.
>>
>>> memset(ring_mem, 0, vq->vq_ring_size);
>>> - vring_init(vr, size, ring_mem, VIRTIO_PCI_VRING_ALIGN);
>>> + vring_init(vq->hw, vr, size, ring_mem, VIRTIO_PCI_VRING_ALIGN);
>>> +
>>> vq->vq_used_cons_idx = 0;
>>> vq->vq_desc_head_idx = 0;
>>> vq->vq_avail_idx = 0;
>>> vq->vq_desc_tail_idx = (uint16_t)(vq->vq_nentries - 1);
>>> vq->vq_free_cnt = vq->vq_nentries;
>>> memset(vq->vq_descx, 0, sizeof(struct vq_desc_extra) *
>>> vq->vq_nentries);
>>> -
>>> - vring_desc_init(vr->desc, size);
>>> + if (vtpci_packed_queue(vq->hw)) {
>>> + if (!vtpci_with_feature(vq->hw, VIRTIO_F_IN_ORDER))
>>> + vring_desc_init_packed(vq, size);
>>
>> Don't we need to initialize descs in IN-ORDER case?
>
> Yes, I missed this here. I'll fix it.
>>
>>> + } else {
>>> + vring_desc_init_split(vr->desc, size);
>>> + }
>>>
>>> /*
>>> * Disable device(host) interrupting guest
>>> @@ -386,7 +388,7 @@ virtio_init_queue(struct rte_eth_dev *dev,
>>> uint16_t vtpci_queue_idx)
>>> /*
>>> * Reserve a memzone for vring elements
>>> */
>>> - size = vring_size(vq_size, VIRTIO_PCI_VRING_ALIGN);
>>> + size = vring_size(hw, vq_size, VIRTIO_PCI_VRING_ALIGN);
>>> vq->vq_ring_size = RTE_ALIGN_CEIL(size, VIRTIO_PCI_VRING_ALIGN);
>>> PMD_INIT_LOG(DEBUG, "vring_size: %d, rounded_vring_size: %d",
>>> size, vq->vq_ring_size);
>>> @@ -489,7 +491,7 @@ virtio_init_queue(struct rte_eth_dev *dev,
>>> uint16_t vtpci_queue_idx)
>>> for (i = 0; i < vq_size; i++) {
>>> struct vring_desc *start_dp = txr[i].tx_indir;
>>>
>>> - vring_desc_init(start_dp, RTE_DIM(txr[i].tx_indir));
>>> + vring_desc_init_split(start_dp, RTE_DIM(txr[i].tx_indir));
>>>
>>> /* first indirect descriptor is always the tx header */
>>> start_dp->addr = txvq->virtio_net_hdr_mem
>>> diff --git a/drivers/net/virtio/virtio_pci.h
>>> b/drivers/net/virtio/virtio_pci.h
>>> index 58fdd3d45..c487d2be0 100644
>>> --- a/drivers/net/virtio/virtio_pci.h
>>> +++ b/drivers/net/virtio/virtio_pci.h
>>> @@ -113,6 +113,7 @@ struct virtnet_ctl;
>>>
>>> #define VIRTIO_F_VERSION_1 32
>>> #define VIRTIO_F_IOMMU_PLATFORM 33
>>> +#define VIRTIO_F_RING_PACKED 34
>>>
>>> /*
>>> * Some VirtIO feature bits (currently bits 28 through 31) are
>>> @@ -314,6 +315,12 @@ vtpci_with_feature(struct virtio_hw *hw,
>>> uint64_t bit)
>>> return (hw->guest_features & (1ULL << bit)) != 0;
>>> }
>>>
>>> +static inline int
>>> +vtpci_packed_queue(struct virtio_hw *hw)
>>> +{
>>> + return vtpci_with_feature(hw, VIRTIO_F_RING_PACKED);
>>> +}
>>> +
>>> /*
>>> * Function declaration from virtio_pci.c
>>> */
>>> diff --git a/drivers/net/virtio/virtio_ring.h
>>> b/drivers/net/virtio/virtio_ring.h
>>> index 9e3c2a015..e65da100b 100644
>>> --- a/drivers/net/virtio/virtio_ring.h
>>> +++ b/drivers/net/virtio/virtio_ring.h
>>> @@ -54,11 +54,38 @@ struct vring_used {
>>> struct vring_used_elem ring[0];
>>> };
>>>
>>> +/* For support of packed virtqueues in Virtio 1.1 the format of
>>> descriptors
>>> + * looks like this.
>>> + */
>>> +struct vring_desc_packed {
>>> + uint64_t addr;
>>> + uint32_t len;
>>> + uint16_t index;
>>
>> s/index/id/
>
> ok
>>
>>> + uint16_t flags;
>>> +} __attribute__ ((aligned(16)));
>>
>> We don't need the "__attribute__ ((aligned(16)))" in descriptor
>> structure's definition.
>
> Someone requested this in an earlier round of reviews, but I'll
> check again.
>
>>
>>> +
>>> +#define RING_EVENT_FLAGS_ENABLE 0x0
>>> +#define RING_EVENT_FLAGS_DISABLE 0x1
>>> +#define RING_EVENT_FLAGS_DESC 0x2
>>
>> Please add an empty line here.
>
> ok
>
>>
>>> +struct vring_packed_desc_event {
>>> + uint16_t desc_event_off_wrap;
>>> + uint16_t desc_event_flags;
>>> +};
>>> +
>>> struct vring {
>>> unsigned int num;
>>> - struct vring_desc *desc;
>>> - struct vring_avail *avail;
>>> - struct vring_used *used;
>>> + union {
>>> + struct vring_desc_packed *desc_packed;
>>> + struct vring_desc *desc;
>>> + };
>>> + union {
>>> + struct vring_avail *avail;
>>> + struct vring_packed_desc_event *driver_event;
>>> + };
>>> + union {
>>> + struct vring_used *used;
>>> + struct vring_packed_desc_event *device_event;
>>> + };
>>> };
>>>
>>
>> We should define a new `vring` structure for packed ring.
>
> I think it was requested to have it as a union before, but I can do
> it.
I guess oyu vcan have a union between struct vring and struct
vring_packed?
> Thanks for your review!
>
> regards,
> Jens
>>
>> Thanks
next prev parent reply other threads:[~2018-10-25 13:51 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-24 14:32 [dpdk-dev] [PATCH v9 0/8] implement packed virtqueues Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 1/8] net/virtio: vring init for packed queues Jens Freimann
2018-10-25 9:21 ` Tiwei Bie
2018-10-25 13:48 ` Jens Freimann
2018-10-25 13:51 ` Maxime Coquelin [this message]
2018-10-25 15:12 ` Jens Freimann
2018-10-26 6:01 ` Tiwei Bie
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 2/8] net/virtio: add packed virtqueue defines Jens Freimann
2018-10-25 9:22 ` Tiwei Bie
2018-10-25 14:41 ` Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 3/8] net/virtio: add packed virtqueue helpers Jens Freimann
2018-10-25 9:23 ` Tiwei Bie
2018-10-25 14:40 ` Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 4/8] net/virtio: dump packed virtqueue data Jens Freimann
2018-10-25 9:25 ` Tiwei Bie
2018-10-25 13:54 ` Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 5/8] net/virtio: implement transmit path for packed queues Jens Freimann
2018-10-24 16:55 ` Maxime Coquelin
2018-10-25 10:59 ` Jens Freimann
2018-10-25 9:31 ` Tiwei Bie
2018-10-25 13:19 ` Jens Freimann
2018-10-25 10:46 ` Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 6/8] net/virtio: implement receive " Jens Freimann
2018-10-25 9:39 ` Tiwei Bie
2018-10-25 13:54 ` Jens Freimann
2018-10-26 5:43 ` Tiwei Bie
2018-10-31 12:46 ` Jens Freimann
2018-10-25 10:53 ` Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 7/8] net/virtio: add virtio send command packed queue support Jens Freimann
2018-10-24 14:32 ` [dpdk-dev] [PATCH v9 8/8] net/virtio: enable packed virtqueues by default Jens Freimann
2018-10-26 6:06 ` Tiwei Bie
2018-11-02 8:58 ` Jens Freimann
2018-10-25 9:44 ` [dpdk-dev] [PATCH v9 0/8] implement packed virtqueues Tiwei Bie
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=931e66a9-5f22-e241-0cbe-4640698543ec@redhat.com \
--to=maxime.coquelin@redhat.com \
--cc=dev@dpdk.org \
--cc=jfreimann@redhat.com \
--cc=tiwei.bie@intel.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).