From: Maxime Coquelin <maxime.coquelin@redhat.com>
To: Marvin Liu <yong.liu@intel.com>,
tiwei.bie@intel.com, Ferruh Yigit <ferruh.yigit@intel.com>
Cc: zhihong.wang@intel.com, dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH v5 8/9] net/virtio: add in-order Rx/Tx into selection
Date: Mon, 2 Jul 2018 17:18:18 +0200 [thread overview]
Message-ID: <b7ff3c52-440f-c0ac-8cbf-c3bc837fa69b@redhat.com> (raw)
In-Reply-To: <b052207a-73b7-a5a4-da85-6fad7783f6f3@redhat.com>
On 07/02/2018 02:41 PM, Maxime Coquelin wrote:
>
>
> On 07/02/2018 01:24 PM, Maxime Coquelin wrote:
>>
>>
>> On 07/02/2018 03:56 PM, Marvin Liu wrote:
>>> After IN_ORDER Rx/Tx paths added, need to update Rx/Tx path selection
>>> logic.
>>>
>>> Rx path select logic: If IN_ORDER and merge-able are enabled will select
>>> IN_ORDER Rx path. If IN_ORDER is enabled, Rx offload and merge-able are
>>> disabled will select simple Rx path. Otherwise will select normal Rx
>>> path.
>>>
>>> Tx path select logic: If IN_ORDER is enabled will select IN_ORDER Tx
>>> path. Otherwise will select default Tx path.
>>>
>>> Signed-off-by: Marvin Liu <yong.liu@intel.com>
>>> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>>>
>>> diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
>>> index 46e292c4d..7c099fb7c 100644
>>> --- a/doc/guides/nics/virtio.rst
>>> +++ b/doc/guides/nics/virtio.rst
>>> @@ -201,7 +201,7 @@ The packet transmission flow is:
>>> Virtio PMD Rx/Tx Callbacks
>>> --------------------------
>>> -Virtio driver has 3 Rx callbacks and 2 Tx callbacks.
>>> +Virtio driver has 4 Rx callbacks and 3 Tx callbacks.
>>> Rx callbacks:
>>> @@ -215,6 +215,9 @@ Rx callbacks:
>>> Vector version without mergeable Rx buffer support, also fixes
>>> the available
>>> ring indexes and uses vector instructions to optimize performance.
>>> +#. ``virtio_recv_mergeable_pkts_inorder``:
>>> + In-order version with mergeable Rx buffer support.
>>> +
>>> Tx callbacks:
>>> #. ``virtio_xmit_pkts``:
>>> @@ -223,6 +226,8 @@ Tx callbacks:
>>> #. ``virtio_xmit_pkts_simple``:
>>> Vector version fixes the available ring indexes to optimize
>>> performance.
>>> +#. ``virtio_xmit_pkts_inorder``:
>>> + In-order version.
>>> By default, the non-vector callbacks are used:
>>> @@ -254,6 +259,12 @@ Example of using the vector version of the
>>> virtio poll mode driver in
>>> testpmd -l 0-2 -n 4 -- -i --tx-offloads=0x0 --rxq=1 --txq=1
>>> --nb-cores=1
>>> +In-order callbacks only work on simulated virtio user vdev.
>>> +
>>> +* For Rx: If mergeable Rx buffers is enabled and in-order is
>>> enabled then
>>> + ``virtio_xmit_pkts_inorder`` is used.
>>> +
>>> +* For Tx: If in-order is enabled then ``virtio_xmit_pkts_inorder``
>>> is used.
>>> Interrupt mode
>>> --------------
>>> diff --git a/drivers/net/virtio/virtio_ethdev.c
>>> b/drivers/net/virtio/virtio_ethdev.c
>>> index df50a571a..df7981ddb 100644
>>> --- a/drivers/net/virtio/virtio_ethdev.c
>>> +++ b/drivers/net/virtio/virtio_ethdev.c
>>> @@ -1320,6 +1320,11 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
>>> PMD_INIT_LOG(INFO, "virtio: using simple Rx path on port %u",
>>> eth_dev->data->port_id);
>>> eth_dev->rx_pkt_burst = virtio_recv_pkts_vec;
>>> + } else if (hw->use_inorder_rx) {
>>> + PMD_INIT_LOG(INFO,
>>> + "virtio: using inorder mergeable buffer Rx path on port
>>> %u",
>>> + eth_dev->data->port_id);
>>> + eth_dev->rx_pkt_burst = &virtio_recv_mergeable_pkts_inorder;
>>> } else if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
>>> PMD_INIT_LOG(INFO,
>>> "virtio: using mergeable buffer Rx path on port %u",
>>> @@ -1335,6 +1340,10 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
>>> PMD_INIT_LOG(INFO, "virtio: using simple Tx path on port %u",
>>> eth_dev->data->port_id);
>>> eth_dev->tx_pkt_burst = virtio_xmit_pkts_simple;
>>> + } else if (hw->use_inorder_tx) {
>>> + PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on port %u",
>>> + eth_dev->data->port_id);
>>> + eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder;
>>> } else {
>>> PMD_INIT_LOG(INFO, "virtio: using standard Tx path on port
>>> %u",
>>> eth_dev->data->port_id);
>>> @@ -1874,20 +1883,27 @@ virtio_dev_configure(struct rte_eth_dev *dev)
>>> hw->use_simple_rx = 1;
>>> hw->use_simple_tx = 1;
>>> + if (vtpci_with_feature(hw, VIRTIO_F_IN_ORDER)) {
>>> + /* Simple Tx not compatible with in-order ring */
>>> + hw->use_inorder_tx = 1;
>>> + hw->use_simple_tx = 0;
>>> + if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
>>> + hw->use_inorder_rx = 1;
>>> + hw->use_simple_rx = 0;
>>> + } else {
>>> + hw->use_inorder_rx = 0;
>>> + if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
>>> + DEV_RX_OFFLOAD_TCP_CKSUM))
>>> + hw->use_simple_rx = 0;
>> It is applied, but I think this is still not good.
>>
>> Simple Rx is set to 1 by default, so if IN_ORDER isn't negotiated,
>> and UDP/TCP_CSUM is enabled, simple Rx keeps being selected.
>>
>> I'll fix it in my series that I'm doing on top.
>
> Actually, after discussion with Ferruh, I fixed it directly in the patch.
Running some more tests, I noticed that it is still broken.
For example the case where host does not support IN_ORDER.
If mergeable buffer is negotiated, the simple path gets selected, which
is wrong.
I fixed that directly in the patch.
Regards,
Maxime
> Thanks,
> Maxime
>
>> Regards,
>> Maxime
>>
>>> + }
>>> + }
>>> +
>>> #if defined RTE_ARCH_ARM64 || defined RTE_ARCH_ARM
>>> if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON)) {
>>> hw->use_simple_rx = 0;
>>> hw->use_simple_tx = 0;
>>> }
>>> #endif
>>> - if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
>>> - hw->use_simple_rx = 0;
>>> - hw->use_simple_tx = 0;
>>> - }
>>> -
>>> - if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
>>> - DEV_RX_OFFLOAD_TCP_CKSUM))
>>> - hw->use_simple_rx = 0;
>>> return 0;
>>> }
>>>
next prev parent reply other threads:[~2018-07-02 15:18 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-02 13:56 [dpdk-dev] [PATCH v5 0/9] support in-order feature Marvin Liu
2018-07-02 8:33 ` Maxime Coquelin
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 1/9] vhost: advertise " Marvin Liu
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 2/9] net/virtio: add in-order feature bit definition Marvin Liu
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 3/9] net/virtio-user: add unsupported features mask Marvin Liu
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 4/9] net/virtio-user: add mrg-rxbuf and in-order vdev parameters Marvin Liu
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 5/9] net/virtio: free in-order descriptors before device start Marvin Liu
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 6/9] net/virtio: extract common part for in-order functions Marvin Liu
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 7/9] net/virtio: support in-order Rx and Tx Marvin Liu
2018-07-02 16:41 ` Ferruh Yigit
2018-07-02 16:52 ` Ferruh Yigit
2018-07-02 16:53 ` Maxime Coquelin
2018-07-02 16:57 ` Ferruh Yigit
2018-07-03 1:36 ` Liu, Yong
2018-07-03 10:36 ` Ferruh Yigit
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 8/9] net/virtio: add in-order Rx/Tx into selection Marvin Liu
2018-07-02 11:24 ` Maxime Coquelin
2018-07-02 12:41 ` Maxime Coquelin
2018-07-02 15:18 ` Maxime Coquelin [this message]
2018-07-03 1:40 ` Liu, Yong
2018-07-02 13:56 ` [dpdk-dev] [PATCH v5 9/9] net/virtio: advertise support in-order feature Marvin 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=b7ff3c52-440f-c0ac-8cbf-c3bc837fa69b@redhat.com \
--to=maxime.coquelin@redhat.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=tiwei.bie@intel.com \
--cc=yong.liu@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).