search for: viommu_init_vqs

...> + > /* IOMMU API */ > > static bool viommu_capable(enum iommu_cap cap) > @@ -938,19 +1018,44 @@ static struct iommu_ops viommu_ops = { > .put_resv_regions = viommu_put_resv_regions, > }; > > -static int viommu_init_vq(struct viommu_dev *viommu) > +static int viommu_init_vqs(struct viommu_dev *viommu) > { > struct virtio_device *vdev = dev_to_virtio(viommu->dev); > - const char *name = "request"; > - void *ret; > + const char *names[] = { "request", "event" }; > + vq_callback_t *callbacks[] = { > + NULL, /* No a...

...rr(viommu->dev, "kick failed\n"); +} + /* IOMMU API */ static bool viommu_capable(enum iommu_cap cap) @@ -934,19 +1015,44 @@ static struct iommu_ops viommu_ops = { .put_resv_regions = viommu_put_resv_regions, }; -static int viommu_init_vq(struct viommu_dev *viommu) +static int viommu_init_vqs(struct viommu_dev *viommu) { struct virtio_device *vdev = dev_to_virtio(viommu->dev); - const char *name = "request"; - void *ret; + const char *names[] = { "request", "event" }; + vq_callback_t *callbacks[] = { + NULL, /* No async requests */ + viommu_event_ha...

...rr(viommu->dev, "kick failed\n"); +} + /* IOMMU API */ static bool viommu_capable(enum iommu_cap cap) @@ -938,19 +1018,44 @@ static struct iommu_ops viommu_ops = { .put_resv_regions = viommu_put_resv_regions, }; -static int viommu_init_vq(struct viommu_dev *viommu) +static int viommu_init_vqs(struct viommu_dev *viommu) { struct virtio_device *vdev = dev_to_virtio(viommu->dev); - const char *name = "request"; - void *ret; + const char *names[] = { "request", "event" }; + vq_callback_t *callbacks[] = { + NULL, /* No async requests */ + viommu_event_ha...

Implement the virtio-iommu driver following version 0.5 of the specification [1]. Previous version of this code was sent back in April [2], implementing the first public RFC. Since then there has been lots of progress and discussion on the specification side, and I think the driver is in a good shape now. The reason patches 1-3 are only RFC is that I'm waiting on feedback from the Virtio TC

Implement the virtio-iommu driver following version 0.5 of the specification [1]. Previous version of this code was sent back in April [2], implementing the first public RFC. Since then there has been lots of progress and discussion on the specification side, and I think the driver is in a good shape now. The reason patches 1-3 are only RFC is that I'm waiting on feedback from the Virtio TC

...gt;dev, "could not add event buffer\n"); + } + + if (!virtqueue_kick(vq)) + dev_err(viommu->dev, "kick failed\n"); +} + /* IOMMU API */ static struct iommu_domain *viommu_domain_alloc(unsigned type) @@ -887,16 +961,35 @@ static struct iommu_ops viommu_ops = { static int viommu_init_vqs(struct viommu_dev *viommu) { struct virtio_device *vdev = dev_to_virtio(viommu->dev); - const char *name = "request"; - void *ret; + const char *names[] = { "request", "event" }; + vq_callback_t *callbacks[] = { + NULL, /* No async requests */ + viommu_event_ha...

Implement the virtio-iommu driver following version 0.6 of the specification [1]. Previous version, RFCv2, was sent in November [2]. This version addresses Eric's comments and changes the device number. (Since last week I also tested and fixed the probe/release functions, they now use devm properly.) I did not include ACPI support because the next IORT specifications isn't ready yet (even

Implement the virtio-iommu driver following version 0.6 of the specification [1]. Previous version, RFCv2, was sent in November [2]. This version addresses Eric's comments and changes the device number. (Since last week I also tested and fixed the probe/release functions, they now use devm properly.) I did not include ACPI support because the next IORT specifications isn't ready yet (even

....add_device = viommu_add_device, > + .remove_device = viommu_remove_device, > + .device_group = viommu_device_group, > + .get_resv_regions = viommu_get_resv_regions, > + .put_resv_regions = viommu_put_resv_regions, > + .of_xlate = viommu_of_xlate, > +}; > + > +static int viommu_init_vqs(struct viommu_dev *viommu) > +{ > + struct virtio_device *vdev = dev_to_virtio(viommu->dev); > + const char *name = "request"; > + void *ret; > + > + ret = virtio_find_single_vq(vdev, NULL, name); > + if (IS_ERR(ret)) { > + dev_err(viommu->dev, "cannot...

....add_device = viommu_add_device, > + .remove_device = viommu_remove_device, > + .device_group = viommu_device_group, > + .get_resv_regions = viommu_get_resv_regions, > + .put_resv_regions = viommu_put_resv_regions, > + .of_xlate = viommu_of_xlate, > +}; > + > +static int viommu_init_vqs(struct viommu_dev *viommu) > +{ > + struct virtio_device *vdev = dev_to_virtio(viommu->dev); > + const char *name = "request"; > + void *ret; > + > + ret = virtio_find_single_vq(vdev, NULL, name); > + if (IS_ERR(ret)) { > + dev_err(viommu->dev, "cannot...

Implement the virtio-iommu driver, following specification v0.9 [1]. Since v4 [2] I fixed the issues reported by Eric, and added Reviewed-by from Eric and Rob. Thanks! I changed the specification to fix one inconsistency discussed in v4. That the device fills the probe buffer with zeroes is now a "SHOULD" instead of a "MAY", since it's the only way for the driver to know

Implement the virtio-iommu driver, following specification v0.9 [1]. Since v4 [2] I fixed the issues reported by Eric, and added Reviewed-by from Eric and Rob. Thanks! I changed the specification to fix one inconsistency discussed in v4. That the device fills the probe buffer with zeroes is now a "SHOULD" instead of a "MAY", since it's the only way for the driver to know

..., + .iotlb_sync = viommu_iotlb_sync, + .add_device = viommu_add_device, + .remove_device = viommu_remove_device, + .device_group = viommu_device_group, + .get_resv_regions = viommu_get_resv_regions, + .put_resv_regions = viommu_put_resv_regions, + .of_xlate = viommu_of_xlate, +}; + +static int viommu_init_vqs(struct viommu_dev *viommu) +{ + struct virtio_device *vdev = dev_to_virtio(viommu->dev); + const char *name = "request"; + void *ret; + + ret = virtio_find_single_vq(vdev, NULL, name); + if (IS_ERR(ret)) { + dev_err(viommu->dev, "cannot find VQ\n"); + return PTR_ERR(ret)...

..., + .iotlb_sync = viommu_iotlb_sync, + .add_device = viommu_add_device, + .remove_device = viommu_remove_device, + .device_group = viommu_device_group, + .get_resv_regions = viommu_get_resv_regions, + .put_resv_regions = viommu_put_resv_regions, + .of_xlate = viommu_of_xlate, +}; + +static int viommu_init_vqs(struct viommu_dev *viommu) +{ + struct virtio_device *vdev = dev_to_virtio(viommu->dev); + const char *name = "request"; + void *ret; + + ret = virtio_find_single_vq(vdev, NULL, name); + if (IS_ERR(ret)) { + dev_err(viommu->dev, "cannot find VQ\n"); + return PTR_ERR(ret)...

..., + .iotlb_sync = viommu_iotlb_sync, + .add_device = viommu_add_device, + .remove_device = viommu_remove_device, + .device_group = viommu_device_group, + .get_resv_regions = viommu_get_resv_regions, + .put_resv_regions = viommu_put_resv_regions, + .of_xlate = viommu_of_xlate, +}; + +static int viommu_init_vqs(struct viommu_dev *viommu) +{ + struct virtio_device *vdev = dev_to_virtio(viommu->dev); + const char *name = "request"; + void *ret; + + ret = virtio_find_single_vq(vdev, NULL, name); + if (IS_ERR(ret)) { + dev_err(viommu->dev, "cannot find VQ\n"); + return PTR_ERR(ret)...

..., + .iotlb_sync = viommu_iotlb_sync, + .add_device = viommu_add_device, + .remove_device = viommu_remove_device, + .device_group = viommu_device_group, + .get_resv_regions = viommu_get_resv_regions, + .put_resv_regions = viommu_put_resv_regions, + .of_xlate = viommu_of_xlate, +}; + +static int viommu_init_vqs(struct viommu_dev *viommu) +{ + struct virtio_device *vdev = dev_to_virtio(viommu->dev); + const char *name = "request"; + void *ret; + + ret = virtio_find_single_vq(vdev, NULL, name); + if (IS_ERR(ret)) { + dev_err(viommu->dev, "cannot find VQ\n"); + return PTR_ERR(ret)...

The virtio IOMMU is a para-virtualized device, allowing to send IOMMU requests such as map/unmap over virtio-mmio transport without emulating page tables. This implementation handles ATTACH, DETACH, MAP and UNMAP requests. The bulk of the code transforms calls coming from the IOMMU API into corresponding virtio requests. Mappings are kept in an interval tree instead of page tables.

Implement the virtio-iommu driver, following specification v0.8 [1]. Changes since v3 [2]: * Rebase onto v4.20-rc2. Patch 3 now touches drivers/of/base.c instead of drivers/pci/of.c, since the map_rid() function has moved. * Removed the request timeout, that depended on DEBUG. * Other small fixes addressing comments on v3. You can find Linux driver and kvmtool device on my virtio-iommu/v0.8.1

..., + .iotlb_sync = viommu_iotlb_sync, + .add_device = viommu_add_device, + .remove_device = viommu_remove_device, + .device_group = viommu_device_group, + .get_resv_regions = viommu_get_resv_regions, + .put_resv_regions = viommu_put_resv_regions, + .of_xlate = viommu_of_xlate, +}; + +static int viommu_init_vqs(struct viommu_dev *viommu) +{ + struct virtio_device *vdev = dev_to_virtio(viommu->dev); + const char *name = "request"; + void *ret; + + ret = virtio_find_single_vq(vdev, NULL, name); + if (IS_ERR(ret)) { + dev_err(viommu->dev, "cannot find VQ\n"); + return PTR_ERR(ret)...

..., >> + .remove_device = viommu_remove_device, >> + .device_group = viommu_device_group, >> + .get_resv_regions = viommu_get_resv_regions, >> + .put_resv_regions = viommu_put_resv_regions, >> + .of_xlate = viommu_of_xlate, >> +}; >> + >> +static int viommu_init_vqs(struct viommu_dev *viommu) >> +{ >> + struct virtio_device *vdev = dev_to_virtio(viommu->dev); >> + const char *name = "request"; >> + void *ret; >> + >> + ret = virtio_find_single_vq(vdev, NULL, name); >> + if (IS_ERR(ret)) { >> + dev_e...