On Tue, Feb 3, 2026 at 8:48 PM Michael S. Tsirkin <[email protected]> wrote:
>
> On Fri, Sep 23, 2022 at 11:36:38AM +0800, Jason Wang wrote:
> > On Fri, Sep 23, 2022 at 11:33 AM 陈浩 <[email protected]> wrote:
> > >
> > >
> > > On 2022/9/22 17:56, Michael S. Tsirkin wrote:
> > > > On Thu, Sep 22, 2022 at 09:34:41AM +0800, Jason Wang wrote:
> > > >> On Thu, Sep 22, 2022 at 1:58 AM Raphael Norwitz
> > > >> <[email protected]> wrote:
> > > >>> If I read your response on the other thread correctly, this change is
> > > >>> intended
> > > >>>
> > > >>> to prioritize the MAC address exposed by DPDK over the one provided
> > > >>> by the
> > > >>>
> > > >>> QEMU command line? Sounds reasonable in principle, but I would get
> > > >>> confirmation
> > > >>>
> > > >>> from vDPA/vhost-net maintainers.
> > > >> I think the best way is to (and it seems easier)
> > > >>
> > > >> 1) have the management layer to make sure the mac came from cli
> > > >> matches the underlayer vDPA
> > > >> 2) having a sanity check and fail the device initialization if they
> > > >> don't match
> > > >>
> > > >> Thanks
> > > > I think I agree, we should not overwrite user supplied arguments.
> > > > I think the case where we might want to get the mac from VDPA
> > > > and use it is when it's been assigned randomly as opposed to coming
> > > > from cli.
> > > Yes, when the cli passes in a random mac address that is inconsistent
> > > with the vdpa device, qemu still starts the virtual machine as usual,
> > > but in this case, the qemu and vdpa environments do not work correctly.
> > > So I want to get mac address directly from vdpa device instead of cli.
> >
> > Let's teach the management to do that. It has a lot of advantages, more
> > below.
>
>
> No, making command line friendlier is a good idea.
> the random address from cli is just that, random,
> so why not take it from backend?
>
Yes, we have a vdpa fix recently that allows changing the MAC address
via the vdpa tool before userspace loads. I wonder maybe you can
leverage that.
https://patchew.org/linux/[email protected]/
thanks
cindy
>
> > Not sure for the DPDK case, but kernel vDPA allows the mgmt to
> > provision the vDPA with a mac address which could be used in this
> > case.
> >
> > >
> > > Having a sanity check and fail the device initialization if they don't
> > > match is a good idea, but it seems more convenient to directly overwrite
> > > the cli mac address.
> >
> > This will confuse the management where it may do a lot of mac related
> > setups.
> >
> > Thanks
>
> i do not think any management relies on the random mac hack.
>
> > > >
> > > >
> > > >>>
> > > >>>
> > > >>> That said the way you’re hacking the vhost-user code breaks a
> > > >>> valuable check for
> > > >>>
> > > >>> bad vhost-user-blk backends. I would suggest finding another
> > > >>> implementation which
> > > >>>
> > > >>> does not regress functionality for other device types.
> > > >>>
> > > >>>
> > > >>>
> > > >>>
> > > >>>
> > > >>>> From: Hao Chen <[email protected]>
> > > >>>> When use dpdk-vdpa tests vdpa device. You need to specify the mac
> > > >>>> address to
> > > >>>> start the virtual machine through libvirt or qemu, but now, the
> > > >>>> libvirt or
> > > >>>> qemu can call dpdk vdpa vendor driver's ops .get_config through
> > > >>>> vhost_net_get_config
> > > >>>> to get the mac address of the vdpa hardware without manual
> > > >>>> configuration.
> > > >>>> v1->v2:
> > > >>>> Only copy ETH_ALEN data of netcfg for some vdpa device such as
> > > >>>> NVIDIA BLUEFIELD DPU(BF2)'s netcfg->status is not right.
> > > >>>> We only need the mac address and don't care about the status field.
> > > >>>> Signed-off-by: Hao Chen <[email protected]>
> > > >>>> ---
> > > >>>> hw/block/vhost-user-blk.c | 1 -
> > > >>>> hw/net/virtio-net.c | 7 +++++++
> > > >>>> hw/virtio/vhost-user.c | 19 -------------------
> > > >>>> 3 files changed, 7 insertions(+), 20 deletions(-)
> > > >>>> diff --git a/hw/block/vhost-user-blk.c b/hw/block/vhost-user-blk.c
> > > >>>> index 9117222456..5dca4eab09 100644
> > > >>>> --- a/hw/block/vhost-user-blk.c
> > > >>>> +++ b/hw/block/vhost-user-blk.c
> > > >>>> @@ -337,7 +337,6 @@ static int vhost_user_blk_connect(DeviceState
> > > >>>> *dev, Error **errp)
> > > >>>> vhost_dev_set_config_notifier(&s->dev, &blk_ops);
> > > >>>> - s->vhost_user.supports_config = true;
> > > >>>> ret = vhost_dev_init(&s->dev, &s->vhost_user,
> > > >>>> VHOST_BACKEND_TYPE_USER, 0,
> > > >>>> errp);
> > > >>>> if (ret < 0) {
> > > >>>> diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
> > > >>>> index dd0d056fde..90405083b1 100644
> > > >>>> --- a/hw/net/virtio-net.c
> > > >>>> +++ b/hw/net/virtio-net.c
> > > >>>> @@ -166,6 +166,13 @@ static void virtio_net_get_config(VirtIODevice
> > > >>>> *vdev, uint8_t *config)
> > > >>>> }
> > > >>>> memcpy(config, &netcfg, n->config_size);
> > > >>>> }
> > > >>>> + } else if (nc->peer && nc->peer->info->type ==
> > > >>>> NET_CLIENT_DRIVER_VHOST_USER) {
> > > >>>> + ret = vhost_net_get_config(get_vhost_net(nc->peer),
> > > >>>> (uint8_t *)&netcfg,
> > > >>>> + n->config_size);
> > > >>>> + if (ret != -1) {
> > > >>>> + /* Automatically obtain the mac address of the vdpa
> > > >>>> device
> > > >>>> + * when using the dpdk vdpa */
> > > >>>> + memcpy(config, &netcfg, ETH_ALEN);
> > > >>>> }
> > > >>>> }
> > > >>>> diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> > > >>>> index bd24741be8..8b01078249 100644
> > > >>>> --- a/hw/virtio/vhost-user.c
> > > >>>> +++ b/hw/virtio/vhost-user.c
> > > >>>> @@ -2013,8 +2013,6 @@ static int vhost_user_backend_init(struct
> > > >>>> vhost_dev *dev, void *opaque,
> > > >>>> }
> > > >>>> if (virtio_has_feature(features,
> > > >>>> VHOST_USER_F_PROTOCOL_FEATURES)) {
> > > >>>> - bool supports_f_config = vus->supports_config ||
> > > >>>> - (dev->config_ops &&
> > > >>>> dev->config_ops->vhost_dev_config_notifier);
> > > >>>> uint64_t protocol_features;
> > > >>>> dev->backend_features |= 1ULL <<
> > > >>>> VHOST_USER_F_PROTOCOL_FEATURES;
> > > >>>> @@ -2033,23 +2031,6 @@ static int vhost_user_backend_init(struct
> > > >>>> vhost_dev *dev, void *opaque,
> > > >>>> */
> > > >>>> protocol_features &= VHOST_USER_PROTOCOL_FEATURE_MASK;
> > > >>>> - if (supports_f_config) {
> > > >>>> - if (!virtio_has_feature(protocol_features,
> > > >>>> - VHOST_USER_PROTOCOL_F_CONFIG)) {
> > > >>>> - error_setg(errp, "vhost-user device expecting "
> > > >>>> - "VHOST_USER_PROTOCOL_F_CONFIG but the
> > > >>>> vhost-user backend does "
> > > >>>> - "not support it.");
> > > >>>> - return -EPROTO;
> > > >>>> - }
> > > >>>> - } else {
> > > >>>> - if (virtio_has_feature(protocol_features,
> > > >>>> - VHOST_USER_PROTOCOL_F_CONFIG)) {
> > > >>>> - warn_reportf_err(*errp, "vhost-user backend
> > > >>>> supports "
> > > >>>> - "VHOST_USER_PROTOCOL_F_CONFIG but
> > > >>>> QEMU does not.");
> > > >>>> - protocol_features &= ~(1ULL <<
> > > >>>> VHOST_USER_PROTOCOL_F_CONFIG);
> > > >>>> - }
> > > >>>> - }
> > > >>>> -
> > > >>>> /* final set of protocol features */
> > > >>>> dev->protocol_features = protocol_features;
> > > >>>> err = vhost_user_set_protocol_features(dev,
> > > >>>> dev->protocol_features);
> > > >>>> --
> > > >>>> 2.27.0
> > > >>>
> > > >
> > >
>
