From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
To: Maxime Coquelin <maxime.coquelin@redhat.com>,
Chenbo Xia <chenbo.xia@intel.com>,
Ivan Dyukov <i.dyukov@samsung.com>
Cc: dev@dpdk.org, Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>,
stable@dpdk.org
Subject: Re: [dpdk-stable] [PATCH] net/virtio: fix link update in speed feature case
Date: Fri, 29 Oct 2021 13:20:08 +0300 [thread overview]
Message-ID: <9f36c830-1e98-7c6a-feff-1c8c0b4b3902@oktetlabs.ru> (raw)
In-Reply-To: <6b1ce80a-25cd-4d6b-18a2-ccc3b51be7ad@redhat.com>
On 10/29/21 12:42 PM, Maxime Coquelin wrote:
>
>
> On 10/22/21 15:17, Andrew Rybchenko wrote:
>> From: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
>>
>> Link update callback reports speed/duplex based on data
>> filled on device initialization. This is wrong in case of
>> VIRTIO_NET_F_SPEED_DUPLEX is negotiated since link could
>> be down at this time. Fix this function to actually
>> update the HW data in this case with respect to the fact
>> that specifying speed via devarg is a highest priority.
>>
>> Fixes: 1357b4b36246 ("net/virtio: support Virtio link speed feature")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
>> Signed-off-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
>> ---
>> drivers/net/virtio/virtio.h | 5 ++++
>> drivers/net/virtio/virtio_ethdev.c | 47 +++++++++++++++++++++---------
>> 2 files changed, 39 insertions(+), 13 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
>> index e78b2e429e..4fd1427375 100644
>> --- a/drivers/net/virtio/virtio.h
>> +++ b/drivers/net/virtio/virtio.h
>> @@ -178,6 +178,11 @@ struct virtio_hw {
>> uint16_t port_id;
>> uint8_t mac_addr[RTE_ETHER_ADDR_LEN];
>> uint32_t speed; /* link speed in MB */
>> + /*
>> + * Speed is specified via 'speed' devarg or
>> + * negotiated via VIRTIO_NET_F_SPEED_DUPLEX
>> + */
>> + bool get_speed_via_feat;
>
> For better packing of this struct, it may be better to place it just
> before the speed field as it already has a 2 bytes hole above.
>
>> uint8_t duplex;
>> uint8_t intr_lsc;
>> uint16_t max_mtu;
>> diff --git a/drivers/net/virtio/virtio_ethdev.c
>> b/drivers/net/virtio/virtio_ethdev.c
>> index 4001368bc4..3d80f664b3 100644
>> --- a/drivers/net/virtio/virtio_ethdev.c
>> +++ b/drivers/net/virtio/virtio_ethdev.c
>> @@ -1777,6 +1777,32 @@ virtio_configure_intr(struct rte_eth_dev *dev)
>> return 0;
>> }
>> +
>> +static void
>> +virtio_get_speed_duplex(struct rte_eth_dev *eth_dev,
>> + struct rte_eth_link *link)
>> +{
>> + struct virtio_hw *hw = eth_dev->data->dev_private;
>> + struct virtio_net_config *config;
>> + struct virtio_net_config local_config;
>> +
>> + config = &local_config;
>> + virtio_read_dev_config(hw,
>> + offsetof(struct virtio_net_config, speed),
>> + &config->speed, sizeof(config->speed));
>> + virtio_read_dev_config(hw,
>> + offsetof(struct virtio_net_config, duplex),
>> + &config->duplex, sizeof(config->duplex));
>> + hw->speed = config->speed;
>> + hw->duplex = config->duplex;
>> + if (link != NULL) {
>> + link->link_duplex = hw->duplex;
>> + link->link_speed = hw->speed;
>> + }
>> + PMD_INIT_LOG(DEBUG, "link speed = %d, duplex = %d",
>> + hw->speed, hw->duplex);
>> +}
>> +
>> #define DUPLEX_UNKNOWN 0xff
>> /* reset device and renegotiate features if needed */
>> static int
>> @@ -1830,19 +1856,10 @@ virtio_init_device(struct rte_eth_dev
>> *eth_dev, uint64_t req_features)
>> hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
>> hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
>> - if (hw->speed == ETH_SPEED_NUM_UNKNOWN) {
>> - if (virtio_with_feature(hw, VIRTIO_NET_F_SPEED_DUPLEX)) {
>> - config = &local_config;
>> - virtio_read_dev_config(hw,
>> - offsetof(struct virtio_net_config, speed),
>> - &config->speed, sizeof(config->speed));
>> - virtio_read_dev_config(hw,
>> - offsetof(struct virtio_net_config, duplex),
>> - &config->duplex, sizeof(config->duplex));
>> - hw->speed = config->speed;
>> - hw->duplex = config->duplex;
>> - }
>> - }
>> + hw->get_speed_via_feat = hw->speed == ETH_SPEED_NUM_UNKNOWN &&
>> + virtio_with_feature(hw, VIRTIO_NET_F_SPEED_DUPLEX);
>> + if (hw->get_speed_via_feat)
>> + virtio_get_speed_duplex(eth_dev, NULL);
>> if (hw->duplex == DUPLEX_UNKNOWN)
>> hw->duplex = ETH_LINK_FULL_DUPLEX;
>> PMD_INIT_LOG(DEBUG, "link speed = %d, duplex = %d",
>> @@ -2554,11 +2571,15 @@ virtio_dev_link_update(struct rte_eth_dev
>> *dev, __rte_unused int wait_to_complet
>> dev->data->port_id);
>> } else {
>> link.link_status = ETH_LINK_UP;
>> + if (hw->get_speed_via_feat)
>> + virtio_get_speed_duplex(dev, &link);
>> PMD_INIT_LOG(DEBUG, "Port %d is up",
>> dev->data->port_id);
>> }
>> } else {
>> link.link_status = ETH_LINK_UP;
>> + if (hw->get_speed_via_feat)
>> + virtio_get_speed_duplex(dev, &link);
>> }
>> return rte_eth_linkstatus_set(dev, &link);
>>
>
> If you agree with the change, I can do it while applying.
Yes, please, do.
>
> Other than that:
> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>
> Thanks,
> Maxime
Many thanks,
Andrew.
next prev parent reply other threads:[~2021-10-29 10:20 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-22 13:17 Andrew Rybchenko
2021-10-29 9:42 ` Maxime Coquelin
2021-10-29 10:20 ` Andrew Rybchenko [this message]
2021-10-29 10:33 ` 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=9f36c830-1e98-7c6a-feff-1c8c0b4b3902@oktetlabs.ru \
--to=andrew.rybchenko@oktetlabs.ru \
--cc=chenbo.xia@intel.com \
--cc=dev@dpdk.org \
--cc=i.dyukov@samsung.com \
--cc=ivan.ilchenko@oktetlabs.ru \
--cc=maxime.coquelin@redhat.com \
--cc=stable@dpdk.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).