Message ID | 20230112172434.760850-7-eperezma@redhat.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | Dinamycally switch to vhost shadow virtqueues at vdpa net migration | expand |
On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: > > To restore the device at the destination of a live migration we send the > commands through control virtqueue. For a device to read CVQ it must > have received the DRIVER_OK status bit. This probably requires the support from the parent driver and requires some changes or fixes in the parent driver. Some drivers did: parent_set_status(): if (DRIVER_OK) if (queue_enable) write queue_enable to the device Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. > > However this opens a window where the device could start receiving > packets in rx queue 0 before it receives the RSS configuration. To avoid > that, we will not send vring_enable until all configuration is used by > the device. > > As a first step, run vhost_set_vring_ready for all vhost_net backend after > all of them are started (with DRIVER_OK). This code should not affect > vdpa. > > Signed-off-by: Eugenio Pérez <eperezma@redhat.com> > --- > hw/net/vhost_net.c | 17 ++++++++++++----- > 1 file changed, 12 insertions(+), 5 deletions(-) > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c > index c4eecc6f36..3900599465 100644 > --- a/hw/net/vhost_net.c > +++ b/hw/net/vhost_net.c > @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > } else { > peer = qemu_get_peer(ncs, n->max_queue_pairs); > } > + r = vhost_net_start_one(get_vhost_net(peer), dev); > + if (r < 0) { > + goto err_start; > + } > + } > + > + for (int j = 0; j < nvhosts; j++) { > + if (j < data_queue_pairs) { > + peer = qemu_get_peer(ncs, j); > + } else { > + peer = qemu_get_peer(ncs, n->max_queue_pairs); > + } I fail to understand why we need to change the vhost_net layer? This is vhost-vDPA specific, so I wonder if we can limit the changes to e.g vhost_vdpa_dev_start()? Thanks > > if (peer->vring_enable) { > /* restore vring enable state */ > @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > goto err_start; > } > } > - > - r = vhost_net_start_one(get_vhost_net(peer), dev); > - if (r < 0) { > - goto err_start; > - } > } > > return 0; > -- > 2.31.1 >
On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: > > On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: > > > > To restore the device at the destination of a live migration we send the > > commands through control virtqueue. For a device to read CVQ it must > > have received the DRIVER_OK status bit. > > This probably requires the support from the parent driver and requires > some changes or fixes in the parent driver. > > Some drivers did: > > parent_set_status(): > if (DRIVER_OK) > if (queue_enable) > write queue_enable to the device > > Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. > I don't get your point here. No device should start reading CVQ (or any other VQ) without having received DRIVER_OK. Some parent drivers do not support sending the queue enable command after DRIVER_OK, usually because they clean part of the state like the set by set_vring_base. Even vdpa_net_sim needs fixes here. But my understanding is that it should be supported so I consider it a bug. Especially after queue_reset patches. Is that what you mean? > > > > However this opens a window where the device could start receiving > > packets in rx queue 0 before it receives the RSS configuration. To avoid > > that, we will not send vring_enable until all configuration is used by > > the device. > > > > As a first step, run vhost_set_vring_ready for all vhost_net backend after > > all of them are started (with DRIVER_OK). This code should not affect > > vdpa. > > > > Signed-off-by: Eugenio Pérez <eperezma@redhat.com> > > --- > > hw/net/vhost_net.c | 17 ++++++++++++----- > > 1 file changed, 12 insertions(+), 5 deletions(-) > > > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c > > index c4eecc6f36..3900599465 100644 > > --- a/hw/net/vhost_net.c > > +++ b/hw/net/vhost_net.c > > @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > > } else { > > peer = qemu_get_peer(ncs, n->max_queue_pairs); > > } > > + r = vhost_net_start_one(get_vhost_net(peer), dev); > > + if (r < 0) { > > + goto err_start; > > + } > > + } > > + > > + for (int j = 0; j < nvhosts; j++) { > > + if (j < data_queue_pairs) { > > + peer = qemu_get_peer(ncs, j); > > + } else { > > + peer = qemu_get_peer(ncs, n->max_queue_pairs); > > + } > > I fail to understand why we need to change the vhost_net layer? This > is vhost-vDPA specific, so I wonder if we can limit the changes to e.g > vhost_vdpa_dev_start()? > The vhost-net layer explicitly calls vhost_set_vring_enable before vhost_dev_start, and this is exactly the behavior we want to avoid. Even if we make changes to vhost_dev, this change is still needed. And we want to explicitly enable CVQ first, which "only" vhost_net knows which is. To perform that in vhost_vdpa_dev_start would require quirks, involving one or more of: * Ignore vq enable calls if the device is not the CVQ one. How to signal what is the CVQ? Can we trust it will be the last one for all kind of devices? * Enable queues that do not belong to the last vhost_dev from the enable call. * Enable the rest of the queues from the last enable in reverse order. * Intercalate the "net load" callback between enabling the last vhost_vdpa device and enabling the rest of devices. * Add an "enable priority" order? Thanks! > Thanks > > > > > if (peer->vring_enable) { > > /* restore vring enable state */ > > @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > > goto err_start; > > } > > } > > - > > - r = vhost_net_start_one(get_vhost_net(peer), dev); > > - if (r < 0) { > > - goto err_start; > > - } > > } > > > > return 0; > > -- > > 2.31.1 > > >
On Fri, Jan 13, 2023 at 09:19:00AM +0100, Eugenio Perez Martin wrote: >On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: >> >> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: >> > >> > To restore the device at the destination of a live migration we send the >> > commands through control virtqueue. For a device to read CVQ it must >> > have received the DRIVER_OK status bit. >> >> This probably requires the support from the parent driver and requires >> some changes or fixes in the parent driver. >> >> Some drivers did: >> >> parent_set_status(): >> if (DRIVER_OK) >> if (queue_enable) >> write queue_enable to the device >> >> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. >> > >I don't get your point here. No device should start reading CVQ (or >any other VQ) without having received DRIVER_OK. > >Some parent drivers do not support sending the queue enable command >after DRIVER_OK, usually because they clean part of the state like the >set by set_vring_base. Even vdpa_net_sim needs fixes here. > >But my understanding is that it should be supported so I consider it a >bug. Especially after queue_reset patches. Is that what you mean? > >> > >> > However this opens a window where the device could start receiving >> > packets in rx queue 0 before it receives the RSS configuration. To avoid >> > that, we will not send vring_enable until all configuration is used by >> > the device. >> > >> > As a first step, run vhost_set_vring_ready for all vhost_net backend after >> > all of them are started (with DRIVER_OK). This code should not affect >> > vdpa. >> > >> > Signed-off-by: Eugenio Pérez <eperezma@redhat.com> >> > --- >> > hw/net/vhost_net.c | 17 ++++++++++++----- >> > 1 file changed, 12 insertions(+), 5 deletions(-) >> > >> > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c >> > index c4eecc6f36..3900599465 100644 >> > --- a/hw/net/vhost_net.c >> > +++ b/hw/net/vhost_net.c >> > @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >> > } else { >> > peer = qemu_get_peer(ncs, n->max_queue_pairs); >> > } >> > + r = vhost_net_start_one(get_vhost_net(peer), dev); >> > + if (r < 0) { >> > + goto err_start; >> > + } >> > + } >> > + >> > + for (int j = 0; j < nvhosts; j++) { >> > + if (j < data_queue_pairs) { >> > + peer = qemu_get_peer(ncs, j); >> > + } else { >> > + peer = qemu_get_peer(ncs, n->max_queue_pairs); >> > + } >> >> I fail to understand why we need to change the vhost_net layer? This >> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g >> vhost_vdpa_dev_start()? >> > >The vhost-net layer explicitly calls vhost_set_vring_enable before >vhost_dev_start, and this is exactly the behavior we want to avoid. >Even if we make changes to vhost_dev, this change is still needed. I'm working on something similar since I'd like to re-work the following commit we merged just before 7.2 release: 4daa5054c5 vhost: enable vrings in vhost_dev_start() for vhost-user devices vhost-net wasn't the only one who enabled vrings independently, but it was easy enough for others devices to avoid it and enable them in vhost_dev_start(). Do you think can we avoid in some way this special behaviour of vhost-net and enable the vrings in vhost_dev_start? Thanks, Stefano > >And we want to explicitly enable CVQ first, which "only" vhost_net >knows which is. To perform that in vhost_vdpa_dev_start would require >quirks, involving one or more of: >* Ignore vq enable calls if the device is not the CVQ one. How to >signal what is the CVQ? Can we trust it will be the last one for all >kind of devices? >* Enable queues that do not belong to the last vhost_dev from the enable call. >* Enable the rest of the queues from the last enable in reverse order. >* Intercalate the "net load" callback between enabling the last >vhost_vdpa device and enabling the rest of devices. >* Add an "enable priority" order? > >Thanks! > >> Thanks >> >> > >> > if (peer->vring_enable) { >> > /* restore vring enable state */ >> > @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >> > goto err_start; >> > } >> > } >> > - >> > - r = vhost_net_start_one(get_vhost_net(peer), dev); >> > - if (r < 0) { >> > - goto err_start; >> > - } >> > } >> > >> > return 0; >> > -- >> > 2.31.1 >> > >> >
On Fri, Jan 13, 2023 at 10:51 AM Stefano Garzarella <sgarzare@redhat.com> wrote: > > On Fri, Jan 13, 2023 at 09:19:00AM +0100, Eugenio Perez Martin wrote: > >On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: > >> > >> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: > >> > > >> > To restore the device at the destination of a live migration we send the > >> > commands through control virtqueue. For a device to read CVQ it must > >> > have received the DRIVER_OK status bit. > >> > >> This probably requires the support from the parent driver and requires > >> some changes or fixes in the parent driver. > >> > >> Some drivers did: > >> > >> parent_set_status(): > >> if (DRIVER_OK) > >> if (queue_enable) > >> write queue_enable to the device > >> > >> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. > >> > > > >I don't get your point here. No device should start reading CVQ (or > >any other VQ) without having received DRIVER_OK. > > > >Some parent drivers do not support sending the queue enable command > >after DRIVER_OK, usually because they clean part of the state like the > >set by set_vring_base. Even vdpa_net_sim needs fixes here. > > > >But my understanding is that it should be supported so I consider it a > >bug. Especially after queue_reset patches. Is that what you mean? > > > >> > > >> > However this opens a window where the device could start receiving > >> > packets in rx queue 0 before it receives the RSS configuration. To avoid > >> > that, we will not send vring_enable until all configuration is used by > >> > the device. > >> > > >> > As a first step, run vhost_set_vring_ready for all vhost_net backend after > >> > all of them are started (with DRIVER_OK). This code should not affect > >> > vdpa. > >> > > >> > Signed-off-by: Eugenio Pérez <eperezma@redhat.com> > >> > --- > >> > hw/net/vhost_net.c | 17 ++++++++++++----- > >> > 1 file changed, 12 insertions(+), 5 deletions(-) > >> > > >> > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c > >> > index c4eecc6f36..3900599465 100644 > >> > --- a/hw/net/vhost_net.c > >> > +++ b/hw/net/vhost_net.c > >> > @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > >> > } else { > >> > peer = qemu_get_peer(ncs, n->max_queue_pairs); > >> > } > >> > + r = vhost_net_start_one(get_vhost_net(peer), dev); > >> > + if (r < 0) { > >> > + goto err_start; > >> > + } > >> > + } > >> > + > >> > + for (int j = 0; j < nvhosts; j++) { > >> > + if (j < data_queue_pairs) { > >> > + peer = qemu_get_peer(ncs, j); > >> > + } else { > >> > + peer = qemu_get_peer(ncs, n->max_queue_pairs); > >> > + } > >> > >> I fail to understand why we need to change the vhost_net layer? This > >> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g > >> vhost_vdpa_dev_start()? > >> > > > >The vhost-net layer explicitly calls vhost_set_vring_enable before > >vhost_dev_start, and this is exactly the behavior we want to avoid. > >Even if we make changes to vhost_dev, this change is still needed. > > I'm working on something similar since I'd like to re-work the following > commit we merged just before 7.2 release: > 4daa5054c5 vhost: enable vrings in vhost_dev_start() for vhost-user > devices > > vhost-net wasn't the only one who enabled vrings independently, but it > was easy enough for others devices to avoid it and enable them in > vhost_dev_start(). > > Do you think can we avoid in some way this special behaviour of > vhost-net and enable the vrings in vhost_dev_start? > Actually looking forward to it :). If that gets merged before this series, I think we could drop this patch. If I'm not wrong the enable/disable dance is used just by vhost-user at the moment. Maxime, could you give us some hints about the tests to use to check that changes do not introduce regressions in vhost-user? Thanks! > Thanks, > Stefano > > > > >And we want to explicitly enable CVQ first, which "only" vhost_net > >knows which is. To perform that in vhost_vdpa_dev_start would require > >quirks, involving one or more of: > >* Ignore vq enable calls if the device is not the CVQ one. How to > >signal what is the CVQ? Can we trust it will be the last one for all > >kind of devices? > >* Enable queues that do not belong to the last vhost_dev from the enable call. > >* Enable the rest of the queues from the last enable in reverse order. > >* Intercalate the "net load" callback between enabling the last > >vhost_vdpa device and enabling the rest of devices. > >* Add an "enable priority" order? > > > >Thanks! > > > >> Thanks > >> > >> > > >> > if (peer->vring_enable) { > >> > /* restore vring enable state */ > >> > @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > >> > goto err_start; > >> > } > >> > } > >> > - > >> > - r = vhost_net_start_one(get_vhost_net(peer), dev); > >> > - if (r < 0) { > >> > - goto err_start; > >> > - } > >> > } > >> > > >> > return 0; > >> > -- > >> > 2.31.1 > >> > > >> > > >
On Fri, Jan 13, 2023 at 11:03:17AM +0100, Eugenio Perez Martin wrote: >On Fri, Jan 13, 2023 at 10:51 AM Stefano Garzarella <sgarzare@redhat.com> wrote: >> >> On Fri, Jan 13, 2023 at 09:19:00AM +0100, Eugenio Perez Martin wrote: >> >On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: >> >> >> >> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: >> >> > >> >> > To restore the device at the destination of a live migration we send the >> >> > commands through control virtqueue. For a device to read CVQ it must >> >> > have received the DRIVER_OK status bit. >> >> >> >> This probably requires the support from the parent driver and requires >> >> some changes or fixes in the parent driver. >> >> >> >> Some drivers did: >> >> >> >> parent_set_status(): >> >> if (DRIVER_OK) >> >> if (queue_enable) >> >> write queue_enable to the device >> >> >> >> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. >> >> >> > >> >I don't get your point here. No device should start reading CVQ (or >> >any other VQ) without having received DRIVER_OK. >> > >> >Some parent drivers do not support sending the queue enable command >> >after DRIVER_OK, usually because they clean part of the state like the >> >set by set_vring_base. Even vdpa_net_sim needs fixes here. >> > >> >But my understanding is that it should be supported so I consider it a >> >bug. Especially after queue_reset patches. Is that what you mean? >> > >> >> > >> >> > However this opens a window where the device could start receiving >> >> > packets in rx queue 0 before it receives the RSS configuration. To avoid >> >> > that, we will not send vring_enable until all configuration is used by >> >> > the device. >> >> > >> >> > As a first step, run vhost_set_vring_ready for all vhost_net backend after >> >> > all of them are started (with DRIVER_OK). This code should not affect >> >> > vdpa. >> >> > >> >> > Signed-off-by: Eugenio Pérez <eperezma@redhat.com> >> >> > --- >> >> > hw/net/vhost_net.c | 17 ++++++++++++----- >> >> > 1 file changed, 12 insertions(+), 5 deletions(-) >> >> > >> >> > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c >> >> > index c4eecc6f36..3900599465 100644 >> >> > --- a/hw/net/vhost_net.c >> >> > +++ b/hw/net/vhost_net.c >> >> > @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >> >> > } else { >> >> > peer = qemu_get_peer(ncs, n->max_queue_pairs); >> >> > } >> >> > + r = vhost_net_start_one(get_vhost_net(peer), dev); >> >> > + if (r < 0) { >> >> > + goto err_start; >> >> > + } >> >> > + } >> >> > + >> >> > + for (int j = 0; j < nvhosts; j++) { >> >> > + if (j < data_queue_pairs) { >> >> > + peer = qemu_get_peer(ncs, j); >> >> > + } else { >> >> > + peer = qemu_get_peer(ncs, n->max_queue_pairs); >> >> > + } >> >> >> >> I fail to understand why we need to change the vhost_net layer? This >> >> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g >> >> vhost_vdpa_dev_start()? >> >> >> > >> >The vhost-net layer explicitly calls vhost_set_vring_enable before >> >vhost_dev_start, and this is exactly the behavior we want to avoid. >> >Even if we make changes to vhost_dev, this change is still needed. >> >> I'm working on something similar since I'd like to re-work the following >> commit we merged just before 7.2 release: >> 4daa5054c5 vhost: enable vrings in vhost_dev_start() for vhost-user >> devices >> >> vhost-net wasn't the only one who enabled vrings independently, but it >> was easy enough for others devices to avoid it and enable them in >> vhost_dev_start(). >> >> Do you think can we avoid in some way this special behaviour of >> vhost-net and enable the vrings in vhost_dev_start? >> > >Actually looking forward to it :). If that gets merged before this >series, I think we could drop this patch. I hope to send a RFC net week :-) let's see... > >If I'm not wrong the enable/disable dance is used just by vhost-user >at the moment. Yep, I got the same. My doubts are that for vhost-user-net we enable only the first VirtIONet->curr_queue_pairs queue IIUC. While for other devices (and maybe also for vDPA devices) we enable all of them. I need to figure out if it's safe to do this for vhost-user-net as well, otherwise I need to find a way to leave this behavior. > >Maxime, could you give us some hints about the tests to use to check >that changes do not introduce regressions in vhost-user? Yep, any help on how to test is very appreciated. Thanks, Stefano
在 2023/1/13 16:19, Eugenio Perez Martin 写道: > On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: >> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: >>> To restore the device at the destination of a live migration we send the >>> commands through control virtqueue. For a device to read CVQ it must >>> have received the DRIVER_OK status bit. >> This probably requires the support from the parent driver and requires >> some changes or fixes in the parent driver. >> >> Some drivers did: >> >> parent_set_status(): >> if (DRIVER_OK) >> if (queue_enable) >> write queue_enable to the device >> >> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. >> > I don't get your point here. No device should start reading CVQ (or > any other VQ) without having received DRIVER_OK. If I understand the code correctly: For CVQ, we do SET_VRING_ENABLE before DRIVER_OK, that's fine. For datapath_vq, we do SET_VRING_ENABLE after DRIVER_OK, this requires parent driver support (explained above) > > Some parent drivers do not support sending the queue enable command > after DRIVER_OK, usually because they clean part of the state like the > set by set_vring_base. Even vdpa_net_sim needs fixes here. Yes, so the question is: Do we need another backend feature for this? (otherwise thing may break silently) > > But my understanding is that it should be supported so I consider it a > bug. Probably, we need fine some proof in the spec, e.g in 3.1.1: """ 7.Perform device-specific setup, including discovery of virtqueues for the device, optional per-bus setup, reading and possibly writing the device’s virtio configuration space, and population of virtqueues. 8.Set the DRIVER_OK status bit. At this point the device is “live”. """ So if my understanding is correct, "discovery of virtqueues for the device" implies queue_enable here which is expected to be done before DRIVER_OK. But it doesn't say anything regrading to the behaviour of setting queue ready after DRIVER_OK. I'm not sure it's a real bug or not, may Michael and comment on this. > Especially after queue_reset patches. Is that what you mean? We haven't supported queue_reset yet in Qemu. But it allows to write 1 to queue_enable after DRIVER_OK for sure. > >>> However this opens a window where the device could start receiving >>> packets in rx queue 0 before it receives the RSS configuration. To avoid >>> that, we will not send vring_enable until all configuration is used by >>> the device. >>> >>> As a first step, run vhost_set_vring_ready for all vhost_net backend after >>> all of them are started (with DRIVER_OK). This code should not affect >>> vdpa. >>> >>> Signed-off-by: Eugenio Pérez <eperezma@redhat.com> >>> --- >>> hw/net/vhost_net.c | 17 ++++++++++++----- >>> 1 file changed, 12 insertions(+), 5 deletions(-) >>> >>> diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c >>> index c4eecc6f36..3900599465 100644 >>> --- a/hw/net/vhost_net.c >>> +++ b/hw/net/vhost_net.c >>> @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >>> } else { >>> peer = qemu_get_peer(ncs, n->max_queue_pairs); >>> } >>> + r = vhost_net_start_one(get_vhost_net(peer), dev); >>> + if (r < 0) { >>> + goto err_start; >>> + } >>> + } >>> + >>> + for (int j = 0; j < nvhosts; j++) { >>> + if (j < data_queue_pairs) { >>> + peer = qemu_get_peer(ncs, j); >>> + } else { >>> + peer = qemu_get_peer(ncs, n->max_queue_pairs); >>> + } >> I fail to understand why we need to change the vhost_net layer? This >> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g >> vhost_vdpa_dev_start()? >> > The vhost-net layer explicitly calls vhost_set_vring_enable before > vhost_dev_start, and this is exactly the behavior we want to avoid. > Even if we make changes to vhost_dev, this change is still needed. Note that the only user of vhost_set_vring_enable() is vhost-user where the semantic is different: It uses that to changes the number of active queues: static int peer_attach(VirtIONet *n, int index) if (nc->peer->info->type == NET_CLIENT_DRIVER_VHOST_USER) { => vhost_set_vring_enable(nc->peer, 1); } This is not the semantic of vhost-vDPA that tries to be complaint with virtio-spec. So I'm not sure how it can help here. > > And we want to explicitly enable CVQ first, which "only" vhost_net > knows which is. This should be known by net/vhost-vdpa.c. > To perform that in vhost_vdpa_dev_start would require > quirks, involving one or more of: > * Ignore vq enable calls if the device is not the CVQ one. How to > signal what is the CVQ? Can we trust it will be the last one for all > kind of devices? > * Enable queues that do not belong to the last vhost_dev from the enable call. > * Enable the rest of the queues from the last enable in reverse order. > * Intercalate the "net load" callback between enabling the last > vhost_vdpa device and enabling the rest of devices. > * Add an "enable priority" order? Haven't had time in thinking through, but it would be better if we can limit the changes in vhost-vdpa layer. E.g currently the VHOST_VDPA_SET_VRING_ENABLE is done at vhost_dev_start(). Thanks > > Thanks! > >> Thanks >> >>> if (peer->vring_enable) { >>> /* restore vring enable state */ >>> @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >>> goto err_start; >>> } >>> } >>> - >>> - r = vhost_net_start_one(get_vhost_net(peer), dev); >>> - if (r < 0) { >>> - goto err_start; >>> - } >>> } >>> >>> return 0; >>> -- >>> 2.31.1 >>>
On Mon, Jan 16, 2023 at 7:37 AM Jason Wang <jasowang@redhat.com> wrote: > > > 在 2023/1/13 16:19, Eugenio Perez Martin 写道: > > On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: > >> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: > >>> To restore the device at the destination of a live migration we send the > >>> commands through control virtqueue. For a device to read CVQ it must > >>> have received the DRIVER_OK status bit. > >> This probably requires the support from the parent driver and requires > >> some changes or fixes in the parent driver. > >> > >> Some drivers did: > >> > >> parent_set_status(): > >> if (DRIVER_OK) > >> if (queue_enable) > >> write queue_enable to the device > >> > >> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. > >> > > I don't get your point here. No device should start reading CVQ (or > > any other VQ) without having received DRIVER_OK. > > > If I understand the code correctly: > > For CVQ, we do SET_VRING_ENABLE before DRIVER_OK, that's fine. > > For datapath_vq, we do SET_VRING_ENABLE after DRIVER_OK, this requires > parent driver support (explained above) > > > > > > Some parent drivers do not support sending the queue enable command > > after DRIVER_OK, usually because they clean part of the state like the > > set by set_vring_base. Even vdpa_net_sim needs fixes here. > > > Yes, so the question is: > > Do we need another backend feature for this? (otherwise thing may break > silently) > > > > > > But my understanding is that it should be supported so I consider it a > > bug. > > > Probably, we need fine some proof in the spec, e.g in 3.1.1: > > """ > > 7.Perform device-specific setup, including discovery of virtqueues for > the device, optional per-bus setup, reading and possibly writing the > device’s virtio configuration space, and population of virtqueues. > 8.Set the DRIVER_OK status bit. At this point the device is “live”. > > """ > > So if my understanding is correct, "discovery of virtqueues for the > device" implies queue_enable here which is expected to be done before > DRIVER_OK. But it doesn't say anything regrading to the behaviour of > setting queue ready after DRIVER_OK. > > I'm not sure it's a real bug or not, may Michael and comment on this. > Right, input on this topic would be really appreciated. > > > Especially after queue_reset patches. Is that what you mean? > > > We haven't supported queue_reset yet in Qemu. But it allows to write 1 > to queue_enable after DRIVER_OK for sure. > I was not clear, I meant in the emulated device. I'm testing this series with the proposal of _F_STATE. > > > > >>> However this opens a window where the device could start receiving > >>> packets in rx queue 0 before it receives the RSS configuration. To avoid > >>> that, we will not send vring_enable until all configuration is used by > >>> the device. > >>> > >>> As a first step, run vhost_set_vring_ready for all vhost_net backend after > >>> all of them are started (with DRIVER_OK). This code should not affect > >>> vdpa. > >>> > >>> Signed-off-by: Eugenio Pérez <eperezma@redhat.com> > >>> --- > >>> hw/net/vhost_net.c | 17 ++++++++++++----- > >>> 1 file changed, 12 insertions(+), 5 deletions(-) > >>> > >>> diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c > >>> index c4eecc6f36..3900599465 100644 > >>> --- a/hw/net/vhost_net.c > >>> +++ b/hw/net/vhost_net.c > >>> @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > >>> } else { > >>> peer = qemu_get_peer(ncs, n->max_queue_pairs); > >>> } > >>> + r = vhost_net_start_one(get_vhost_net(peer), dev); > >>> + if (r < 0) { > >>> + goto err_start; > >>> + } > >>> + } > >>> + > >>> + for (int j = 0; j < nvhosts; j++) { > >>> + if (j < data_queue_pairs) { > >>> + peer = qemu_get_peer(ncs, j); > >>> + } else { > >>> + peer = qemu_get_peer(ncs, n->max_queue_pairs); > >>> + } > >> I fail to understand why we need to change the vhost_net layer? This > >> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g > >> vhost_vdpa_dev_start()? > >> > > The vhost-net layer explicitly calls vhost_set_vring_enable before > > vhost_dev_start, and this is exactly the behavior we want to avoid. > > Even if we make changes to vhost_dev, this change is still needed. > > > Note that the only user of vhost_set_vring_enable() is vhost-user where > the semantic is different: > > It uses that to changes the number of active queues: > > static int peer_attach(VirtIONet *n, int index) > > if (nc->peer->info->type == NET_CLIENT_DRIVER_VHOST_USER) { > => vhost_set_vring_enable(nc->peer, 1); > } > > This is not the semantic of vhost-vDPA that tries to be complaint with > virtio-spec. So I'm not sure how it can help here. > Right, but previous changes use enable callback to delay the enable of the datapath virtqueues. I'll try to fit the changes in virtio/vhost-vdpa though. Thanks! > > > > > And we want to explicitly enable CVQ first, which "only" vhost_net > > knows which is. > > > This should be known by net/vhost-vdpa.c. > > > > To perform that in vhost_vdpa_dev_start would require > > quirks, involving one or more of: > > * Ignore vq enable calls if the device is not the CVQ one. How to > > signal what is the CVQ? Can we trust it will be the last one for all > > kind of devices? > > * Enable queues that do not belong to the last vhost_dev from the enable call. > > * Enable the rest of the queues from the last enable in reverse order. > > * Intercalate the "net load" callback between enabling the last > > vhost_vdpa device and enabling the rest of devices. > > * Add an "enable priority" order? > > > Haven't had time in thinking through, but it would be better if we can > limit the changes in vhost-vdpa layer. E.g currently the > VHOST_VDPA_SET_VRING_ENABLE is done at vhost_dev_start(). > > Thanks > > > > > > Thanks! > > > >> Thanks > >> > >>> if (peer->vring_enable) { > >>> /* restore vring enable state */ > >>> @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, > >>> goto err_start; > >>> } > >>> } > >>> - > >>> - r = vhost_net_start_one(get_vhost_net(peer), dev); > >>> - if (r < 0) { > >>> - goto err_start; > >>> - } > >>> } > >>> > >>> return 0; > >>> -- > >>> 2.31.1 > >>> >
在 2023/1/17 00:16, Eugenio Perez Martin 写道: > On Mon, Jan 16, 2023 at 7:37 AM Jason Wang <jasowang@redhat.com> wrote: >> >> 在 2023/1/13 16:19, Eugenio Perez Martin 写道: >>> On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: >>>> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: >>>>> To restore the device at the destination of a live migration we send the >>>>> commands through control virtqueue. For a device to read CVQ it must >>>>> have received the DRIVER_OK status bit. >>>> This probably requires the support from the parent driver and requires >>>> some changes or fixes in the parent driver. >>>> >>>> Some drivers did: >>>> >>>> parent_set_status(): >>>> if (DRIVER_OK) >>>> if (queue_enable) >>>> write queue_enable to the device >>>> >>>> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. >>>> >>> I don't get your point here. No device should start reading CVQ (or >>> any other VQ) without having received DRIVER_OK. >> >> If I understand the code correctly: >> >> For CVQ, we do SET_VRING_ENABLE before DRIVER_OK, that's fine. >> >> For datapath_vq, we do SET_VRING_ENABLE after DRIVER_OK, this requires >> parent driver support (explained above) >> >> >>> Some parent drivers do not support sending the queue enable command >>> after DRIVER_OK, usually because they clean part of the state like the >>> set by set_vring_base. Even vdpa_net_sim needs fixes here. >> >> Yes, so the question is: >> >> Do we need another backend feature for this? (otherwise thing may break >> silently) >> >> >>> But my understanding is that it should be supported so I consider it a >>> bug. >> >> Probably, we need fine some proof in the spec, e.g in 3.1.1: >> >> """ >> >> 7.Perform device-specific setup, including discovery of virtqueues for >> the device, optional per-bus setup, reading and possibly writing the >> device’s virtio configuration space, and population of virtqueues. >> 8.Set the DRIVER_OK status bit. At this point the device is “live”. >> >> """ >> >> So if my understanding is correct, "discovery of virtqueues for the >> device" implies queue_enable here which is expected to be done before >> DRIVER_OK. But it doesn't say anything regrading to the behaviour of >> setting queue ready after DRIVER_OK. >> >> I'm not sure it's a real bug or not, may Michael and comment on this. >> > Right, input on this topic would be really appreciated. > >>> Especially after queue_reset patches. Is that what you mean? >> >> We haven't supported queue_reset yet in Qemu. But it allows to write 1 >> to queue_enable after DRIVER_OK for sure. >> > I was not clear, I meant in the emulated device. I'm testing this > series with the proposal of _F_STATE. > >>>>> However this opens a window where the device could start receiving >>>>> packets in rx queue 0 before it receives the RSS configuration. To avoid >>>>> that, we will not send vring_enable until all configuration is used by >>>>> the device. >>>>> >>>>> As a first step, run vhost_set_vring_ready for all vhost_net backend after >>>>> all of them are started (with DRIVER_OK). This code should not affect >>>>> vdpa. >>>>> >>>>> Signed-off-by: Eugenio Pérez <eperezma@redhat.com> >>>>> --- >>>>> hw/net/vhost_net.c | 17 ++++++++++++----- >>>>> 1 file changed, 12 insertions(+), 5 deletions(-) >>>>> >>>>> diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c >>>>> index c4eecc6f36..3900599465 100644 >>>>> --- a/hw/net/vhost_net.c >>>>> +++ b/hw/net/vhost_net.c >>>>> @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >>>>> } else { >>>>> peer = qemu_get_peer(ncs, n->max_queue_pairs); >>>>> } >>>>> + r = vhost_net_start_one(get_vhost_net(peer), dev); >>>>> + if (r < 0) { >>>>> + goto err_start; >>>>> + } >>>>> + } >>>>> + >>>>> + for (int j = 0; j < nvhosts; j++) { >>>>> + if (j < data_queue_pairs) { >>>>> + peer = qemu_get_peer(ncs, j); >>>>> + } else { >>>>> + peer = qemu_get_peer(ncs, n->max_queue_pairs); >>>>> + } >>>> I fail to understand why we need to change the vhost_net layer? This >>>> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g >>>> vhost_vdpa_dev_start()? >>>> >>> The vhost-net layer explicitly calls vhost_set_vring_enable before >>> vhost_dev_start, and this is exactly the behavior we want to avoid. >>> Even if we make changes to vhost_dev, this change is still needed. >> >> Note that the only user of vhost_set_vring_enable() is vhost-user where >> the semantic is different: >> >> It uses that to changes the number of active queues: >> >> static int peer_attach(VirtIONet *n, int index) >> >> if (nc->peer->info->type == NET_CLIENT_DRIVER_VHOST_USER) { >> => vhost_set_vring_enable(nc->peer, 1); >> } >> >> This is not the semantic of vhost-vDPA that tries to be complaint with >> virtio-spec. So I'm not sure how it can help here. >> > Right, but previous changes use enable callback to delay the enable of > the datapath virtqueues. I'll try to fit the changes in > virtio/vhost-vdpa though. This would make things more complicated. As mentioned above, vhost-user's usage of vhost_set_vring_enable() is not spec compliant while the vhost-vDPA VHOST_VDPA_SET_VRING_ENALBE tries to be compliant with the spec. If we tries to mix use that it may result confusion for the readers. Thanks > > Thanks! > >>> And we want to explicitly enable CVQ first, which "only" vhost_net >>> knows which is. >> >> This should be known by net/vhost-vdpa.c. >> >> >>> To perform that in vhost_vdpa_dev_start would require >>> quirks, involving one or more of: >>> * Ignore vq enable calls if the device is not the CVQ one. How to >>> signal what is the CVQ? Can we trust it will be the last one for all >>> kind of devices? >>> * Enable queues that do not belong to the last vhost_dev from the enable call. >>> * Enable the rest of the queues from the last enable in reverse order. >>> * Intercalate the "net load" callback between enabling the last >>> vhost_vdpa device and enabling the rest of devices. >>> * Add an "enable priority" order? >> >> Haven't had time in thinking through, but it would be better if we can >> limit the changes in vhost-vdpa layer. E.g currently the >> VHOST_VDPA_SET_VRING_ENABLE is done at vhost_dev_start(). >> >> Thanks >> >> >>> Thanks! >>> >>>> Thanks >>>> >>>>> if (peer->vring_enable) { >>>>> /* restore vring enable state */ >>>>> @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >>>>> goto err_start; >>>>> } >>>>> } >>>>> - >>>>> - r = vhost_net_start_one(get_vhost_net(peer), dev); >>>>> - if (r < 0) { >>>>> - goto err_start; >>>>> - } >>>>> } >>>>> >>>>> return 0; >>>>> -- >>>>> 2.31.1 >>>>>
Hi Eugenio, On 1/13/23 11:03, Eugenio Perez Martin wrote: > On Fri, Jan 13, 2023 at 10:51 AM Stefano Garzarella <sgarzare@redhat.com> wrote: >> >> On Fri, Jan 13, 2023 at 09:19:00AM +0100, Eugenio Perez Martin wrote: >>> On Fri, Jan 13, 2023 at 5:36 AM Jason Wang <jasowang@redhat.com> wrote: >>>> >>>> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote: >>>>> >>>>> To restore the device at the destination of a live migration we send the >>>>> commands through control virtqueue. For a device to read CVQ it must >>>>> have received the DRIVER_OK status bit. >>>> >>>> This probably requires the support from the parent driver and requires >>>> some changes or fixes in the parent driver. >>>> >>>> Some drivers did: >>>> >>>> parent_set_status(): >>>> if (DRIVER_OK) >>>> if (queue_enable) >>>> write queue_enable to the device >>>> >>>> Examples are IFCVF or even vp_vdpa at least. MLX5 seems to be fine. >>>> >>> >>> I don't get your point here. No device should start reading CVQ (or >>> any other VQ) without having received DRIVER_OK. >>> >>> Some parent drivers do not support sending the queue enable command >>> after DRIVER_OK, usually because they clean part of the state like the >>> set by set_vring_base. Even vdpa_net_sim needs fixes here. >>> >>> But my understanding is that it should be supported so I consider it a >>> bug. Especially after queue_reset patches. Is that what you mean? >>> >>>>> >>>>> However this opens a window where the device could start receiving >>>>> packets in rx queue 0 before it receives the RSS configuration. To avoid >>>>> that, we will not send vring_enable until all configuration is used by >>>>> the device. >>>>> >>>>> As a first step, run vhost_set_vring_ready for all vhost_net backend after >>>>> all of them are started (with DRIVER_OK). This code should not affect >>>>> vdpa. >>>>> >>>>> Signed-off-by: Eugenio Pérez <eperezma@redhat.com> >>>>> --- >>>>> hw/net/vhost_net.c | 17 ++++++++++++----- >>>>> 1 file changed, 12 insertions(+), 5 deletions(-) >>>>> >>>>> diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c >>>>> index c4eecc6f36..3900599465 100644 >>>>> --- a/hw/net/vhost_net.c >>>>> +++ b/hw/net/vhost_net.c >>>>> @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >>>>> } else { >>>>> peer = qemu_get_peer(ncs, n->max_queue_pairs); >>>>> } >>>>> + r = vhost_net_start_one(get_vhost_net(peer), dev); >>>>> + if (r < 0) { >>>>> + goto err_start; >>>>> + } >>>>> + } >>>>> + >>>>> + for (int j = 0; j < nvhosts; j++) { >>>>> + if (j < data_queue_pairs) { >>>>> + peer = qemu_get_peer(ncs, j); >>>>> + } else { >>>>> + peer = qemu_get_peer(ncs, n->max_queue_pairs); >>>>> + } >>>> >>>> I fail to understand why we need to change the vhost_net layer? This >>>> is vhost-vDPA specific, so I wonder if we can limit the changes to e.g >>>> vhost_vdpa_dev_start()? >>>> >>> >>> The vhost-net layer explicitly calls vhost_set_vring_enable before >>> vhost_dev_start, and this is exactly the behavior we want to avoid. >>> Even if we make changes to vhost_dev, this change is still needed. >> >> I'm working on something similar since I'd like to re-work the following >> commit we merged just before 7.2 release: >> 4daa5054c5 vhost: enable vrings in vhost_dev_start() for vhost-user >> devices >> >> vhost-net wasn't the only one who enabled vrings independently, but it >> was easy enough for others devices to avoid it and enable them in >> vhost_dev_start(). >> >> Do you think can we avoid in some way this special behaviour of >> vhost-net and enable the vrings in vhost_dev_start? >> > > Actually looking forward to it :). If that gets merged before this > series, I think we could drop this patch. > > If I'm not wrong the enable/disable dance is used just by vhost-user > at the moment. > > Maxime, could you give us some hints about the tests to use to check > that changes do not introduce regressions in vhost-user? You can use DPDK's testpmd [0] tool with Vhost PMD, e.g.: #single queue pair # dpdk-testpmd -l <CORE IDs> --no-pci --vdev=net_vhost0,iface=/tmp/vhost-user1 -- -i #multiqueue # dpdk-testpmd -l <CORE IDs> --no-pci --vdev=net_vhost0,iface=/tmp/vhost-user1,queues=4 -- -i --rxq=4 --txq=4 [0]: https://doc.dpdk.org/guides/testpmd_app_ug/index.html Maxime > Thanks! > >> Thanks, >> Stefano >> >>> >>> And we want to explicitly enable CVQ first, which "only" vhost_net >>> knows which is. To perform that in vhost_vdpa_dev_start would require >>> quirks, involving one or more of: >>> * Ignore vq enable calls if the device is not the CVQ one. How to >>> signal what is the CVQ? Can we trust it will be the last one for all >>> kind of devices? >>> * Enable queues that do not belong to the last vhost_dev from the enable call. >>> * Enable the rest of the queues from the last enable in reverse order. >>> * Intercalate the "net load" callback between enabling the last >>> vhost_vdpa device and enabling the rest of devices. >>> * Add an "enable priority" order? >>> >>> Thanks! >>> >>>> Thanks >>>> >>>>> >>>>> if (peer->vring_enable) { >>>>> /* restore vring enable state */ >>>>> @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, >>>>> goto err_start; >>>>> } >>>>> } >>>>> - >>>>> - r = vhost_net_start_one(get_vhost_net(peer), dev); >>>>> - if (r < 0) { >>>>> - goto err_start; >>>>> - } >>>>> } >>>>> >>>>> return 0; >>>>> -- >>>>> 2.31.1 >>>>> >>>> >>> >> >
diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c index c4eecc6f36..3900599465 100644 --- a/hw/net/vhost_net.c +++ b/hw/net/vhost_net.c @@ -399,6 +399,18 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, } else { peer = qemu_get_peer(ncs, n->max_queue_pairs); } + r = vhost_net_start_one(get_vhost_net(peer), dev); + if (r < 0) { + goto err_start; + } + } + + for (int j = 0; j < nvhosts; j++) { + if (j < data_queue_pairs) { + peer = qemu_get_peer(ncs, j); + } else { + peer = qemu_get_peer(ncs, n->max_queue_pairs); + } if (peer->vring_enable) { /* restore vring enable state */ @@ -408,11 +420,6 @@ int vhost_net_start(VirtIODevice *dev, NetClientState *ncs, goto err_start; } } - - r = vhost_net_start_one(get_vhost_net(peer), dev); - if (r < 0) { - goto err_start; - } } return 0;
To restore the device at the destination of a live migration we send the commands through control virtqueue. For a device to read CVQ it must have received the DRIVER_OK status bit. However this opens a window where the device could start receiving packets in rx queue 0 before it receives the RSS configuration. To avoid that, we will not send vring_enable until all configuration is used by the device. As a first step, run vhost_set_vring_ready for all vhost_net backend after all of them are started (with DRIVER_OK). This code should not affect vdpa. Signed-off-by: Eugenio Pérez <eperezma@redhat.com> --- hw/net/vhost_net.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)