search for: set_vring_base

On 31.03.20 21:27, Eugenio P?rez wrote: > Vhost did not reset properly the batched descriptors on SET_VRING_BASE > event. Because of that, is possible to return an invalid descriptor to > the guest. > > This series ammend this, resetting them every time backend changes, and > creates a test to assert correct behavior. To do that, they need to > expose a new function in virtio_ring, virtqueu...

On 01.04.20 20:40, Eugenio Perez Martin wrote: > On Wed, Apr 1, 2020 at 9:19 AM Christian Borntraeger > <borntraeger at de.ibm.com> wrote: >> >> On 31.03.20 21:27, Eugenio P?rez wrote: >>> Vhost did not reset properly the batched descriptors on SET_VRING_BASE >>> event. Because of that, is possible to return an invalid descriptor to >>> the guest. >>> >>> This series ammend this, resetting them every time backend changes, and >>> creates a test to assert correct behavior. To do that, they need to >>>...

...idx++) { > > + vq = &m->vqs[idx]; > > + > > + if (!vq->desc || !vq->avail || !vq->used) > > + break; > > + > > + if (ops->set_vq_state(mdev, idx, vq->last_avail_idx)) > > + goto reset; > > > If we do set_vq_state() in SET_VRING_BASE, we won't need this step here. Yeah, I plan to do it in the next version. > > > > + > > + /* > > + * In vhost-mdev, userspace should pass ring addresses > > + * in guest physical addresses when IOMMU is disabled or > > + * IOVAs when IOMMU is enabl...

...idx++) { > > + vq = &m->vqs[idx]; > > + > > + if (!vq->desc || !vq->avail || !vq->used) > > + break; > > + > > + if (ops->set_vq_state(mdev, idx, vq->last_avail_idx)) > > + goto reset; > > > If we do set_vq_state() in SET_VRING_BASE, we won't need this step here. Yeah, I plan to do it in the next version. > > > > + > > + /* > > + * In vhost-mdev, userspace should pass ring addresses > > + * in guest physical addresses when IOMMU is disabled or > > + * IOVAs when IOMMU is enabl...

...> > + > > > > + if (!vq->desc || !vq->avail || !vq->used) > > > > + break; > > > > + > > > > + if (ops->set_vq_state(mdev, idx, vq->last_avail_idx)) > > > > + goto reset; > > > If we do set_vq_state() in SET_VRING_BASE, we won't need this step here. > > Yeah, I plan to do it in the next version. > > > > > > + > > > > + /* > > > > + * In vhost-mdev, userspace should pass ring addresses > > > > + * in guest physical addresses when IOMMU is disabl...

...ports about 20% improvement on PPS (event index is off). More >> testing is ongoing. >> >> Notes for tester: >> >> - Start from this version, vhost need qemu co-operation to work >> correctly. Or you can comment out the packed specific code for >> GET/SET_VRING_BASE. > Do you mean the code in vhost_virtqueue_start/stop? For qemu, probably. > Both Tiwei's and your v3 > work fortunately correctly which should be avoided since the ring should be > definitely different. I don't understand this, you mean reset work? Thanks > > Wei &gt...

...>> + vq = &m->vqs[idx]; >>> + >>> + if (!vq->desc || !vq->avail || !vq->used) >>> + break; >>> + >>> + if (ops->set_vq_state(mdev, idx, vq->last_avail_idx)) >>> + goto reset; >> If we do set_vq_state() in SET_VRING_BASE, we won't need this step here. > Yeah, I plan to do it in the next version. > >>> + >>> + /* >>> + * In vhost-mdev, userspace should pass ring addresses >>> + * in guest physical addresses when IOMMU is disabled or >>> + * IOVAs when IOM...

...io-v1.1-csprd01.html#x1-610007 > > This version were tested with: > - zercopy/datacopy > - mergeable buffer on/off > - TCP stream & virtio-user > > Changes from V2: > - rebase on top of vhost metadata accelreation series > - introduce shadow used ring API > - new SET_VRING_BASE/GET_VRING_BASE that takes care about warp counter > and index for both avail and used > - various twaeaks > > Changes from V1: > - drop uapi patch and use Tiwei's > - split the enablement of packed virtqueue into a separate patch > > Changes from RFC V5: > - save...

...- drop uapi patch and use Tiwei's >> - split the enablement of packed virtqueue into a separate patch >> >> Changes from RFC V5: >> >> - save unnecessary barriers during vhost_add_used_packed_n() >> - more compact math for event idx >> - fix failure of SET_VRING_BASE when avail_wrap_counter is true >> - fix not copy avail_wrap_counter during GET_VRING_BASE >> - introduce SET_VRING_USED_BASE/GET_VRING_USED_BASE for syncing last_used_idx >> - rename used_wrap_counter to last_used_wrap_counter >> - rebase to net-next >> >> Chang...

...- drop uapi patch and use Tiwei's >> - split the enablement of packed virtqueue into a separate patch >> >> Changes from RFC V5: >> >> - save unnecessary barriers during vhost_add_used_packed_n() >> - more compact math for event idx >> - fix failure of SET_VRING_BASE when avail_wrap_counter is true >> - fix not copy avail_wrap_counter during GET_VRING_BASE >> - introduce SET_VRING_USED_BASE/GET_VRING_USED_BASE for syncing last_used_idx >> - rename used_wrap_counter to last_used_wrap_counter >> - rebase to net-next >> >> Chang...

...test? > > Pktgen reports about 20% improvement on PPS (event index is off). More > testing is ongoing. > > Notes for tester: > > - Start from this version, vhost need qemu co-operation to work > correctly. Or you can comment out the packed specific code for > GET/SET_VRING_BASE. Do you mean the code in vhost_virtqueue_start/stop? Both Tiwei's and your v3 work fortunately correctly which should be avoided since the ring should be definitely different. Wei > > Changes from V3: > - Fix math on event idx checking > - Sync last avail wrap counter through GE...

...> Changes from V1: > - drop uapi patch and use Tiwei's > - split the enablement of packed virtqueue into a separate patch > > Changes from RFC V5: > > - save unnecessary barriers during vhost_add_used_packed_n() > - more compact math for event idx > - fix failure of SET_VRING_BASE when avail_wrap_counter is true > - fix not copy avail_wrap_counter during GET_VRING_BASE > - introduce SET_VRING_USED_BASE/GET_VRING_USED_BASE for syncing last_used_idx > - rename used_wrap_counter to last_used_wrap_counter > - rebase to net-next > > Changes from RFC V4: > &...

...> > - split the enablement of packed virtqueue into a separate patch > > > > > > Changes from RFC V5: > > > > > > - save unnecessary barriers during vhost_add_used_packed_n() > > > - more compact math for event idx > > > - fix failure of SET_VRING_BASE when avail_wrap_counter is true > > > - fix not copy avail_wrap_counter during GET_VRING_BASE > > > - introduce SET_VRING_USED_BASE/GET_VRING_USED_BASE for syncing last_used_idx > > > - rename used_wrap_counter to last_used_wrap_counter > > > - rebase to net-next...

...- split the enablement of packed virtqueue into a separate patch >>>> >>>> Changes from RFC V5: >>>> >>>> - save unnecessary barriers during vhost_add_used_packed_n() >>>> - more compact math for event idx >>>> - fix failure of SET_VRING_BASE when avail_wrap_counter is true >>>> - fix not copy avail_wrap_counter during GET_VRING_BASE >>>> - introduce SET_VRING_USED_BASE/GET_VRING_USED_BASE for syncing last_used_idx >>>> - rename used_wrap_counter to last_used_wrap_counter >>>> - rebase to...

...> + > + for (idx = 0; idx < m->nvqs; idx++) { > + vq = &m->vqs[idx]; > + > + if (!vq->desc || !vq->avail || !vq->used) > + break; > + > + if (ops->set_vq_state(mdev, idx, vq->last_avail_idx)) > + goto reset; If we do set_vq_state() in SET_VRING_BASE, we won't need this step here. > + > + /* > + * In vhost-mdev, userspace should pass ring addresses > + * in guest physical addresses when IOMMU is disabled or > + * IOVAs when IOMMU is enabled. > + */ A question here, consider we're using noiommu mode. If gue...

On Fri, Apr 03, 2020 at 06:51:11PM +0200, Eugenio P??rez wrote: > This series add the tests used to validate the "vhost: Reset batched > descriptors on SET_VRING_BASE call" series, with a small change on the > reset code (delete an extra unneded reset on VHOST_SET_VRING_BASE). > > They are based on the tests sent back them, the ones that were not > included (reasons in that thread). This series changes: > > * Delete need to export the ug...

This patch introduces a mdev based hardware vhost backend. This backend is built on top of the same abstraction used in virtio-mdev and provides a generic vhost interface for userspace to accelerate the virtio devices in guest. This backend is implemented as a mdev device driver on top of the same mdev device ops used in virtio-mdev but using a different mdev class id, and it will register the

This patch introduces a mdev based hardware vhost backend. This backend is built on top of the same abstraction used in virtio-mdev and provides a generic vhost interface for userspace to accelerate the virtio devices in guest. This backend is implemented as a mdev device driver on top of the same mdev device ops used in virtio-mdev but using a different mdev class id, and it will register the

...wrote: > On Mon, Apr 13, 2020 at 1:13 PM Michael S. Tsirkin <mst at redhat.com> wrote: > > > > On Fri, Apr 03, 2020 at 06:51:11PM +0200, Eugenio P??rez wrote: > > > This series add the tests used to validate the "vhost: Reset batched > > > descriptors on SET_VRING_BASE call" series, with a small change on the > > > reset code (delete an extra unneded reset on VHOST_SET_VRING_BASE). > > > > > > They are based on the tests sent back them, the ones that were not > > > included (reasons in that thread). This series changes: &g...

...class id and device specific ops (device_ops) > - split device_ops into seperate headers > - drop the mdev_set_dma_ops() > - use device_ops to implement the transport API, then it's not a part > of UAPI any more > - use GFP_ATOMIC in mvnet sample device and other tweaks > - set_vring_base/get_vring_base support for mvnet device > > Jason Wang (8): > vringh: fix copy direction of vringh_iov_push_kern() > mdev: class id support > mdev: bus uevent support > modpost: add support for mdev class id > mdev: introduce device specific ops > mdev: introduc...