similar to: [PATCH v2 3/4] drm/virtio: move mapping teardown to virtio_gpu_cleanup_object()

Stop sending DETACH_BACKING commands, that will happening anyway when releasing resources via UNREF. Handle guest-side cleanup in virtio_gpu_cleanup_object(), called when the host finished processing the UNREF command. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_drv.h | 2 -- drivers/gpu/drm/virtio/virtgpu_object.c | 14 ++++++--

Add new virtio_gpu_cleanup_object() helper function for object cleanup. Wire up callback function for resource unref, do cleanup from callback when we know the host stopped using the resource. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_drv.h | 4 +++- drivers/gpu/drm/virtio/virtgpu_object.c | 19 +++++++++++-----

On Wed, Feb 5, 2020 at 3:00 AM Gerd Hoffmann <kraxel at redhat.com> wrote: > > Add new virtio_gpu_cleanup_object() helper function for object cleanup. > Wire up callback function for resource unref, do cleanup from callback > when we know the host stopped using the resource. > > Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> > --- >

Add new virtio_gpu_cleanup_object() helper function for object cleanup. Wire up callback function for resource unref, do cleanup from callback when we know the host stopped using the resource. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_drv.h | 3 ++- drivers/gpu/drm/virtio/virtgpu_object.c | 19 ++++++++++----

Add new virtio_gpu_cleanup_object() helper function for object cleanup. Wire up callback function for resource unref, do cleanup from callback when we know the host stopped using the resource. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_drv.h | 3 ++- drivers/gpu/drm/virtio/virtgpu_object.c | 19 ++++++++++----

For the series, Reviewed-by: Dave Airlie <airlied at redhat.com> On Wed, 29 Aug 2018 at 22:20, Gerd Hoffmann <kraxel at redhat.com> wrote: > > Use the dma mapping api and properly add iommu mappings for > objects, unless virtio is in iommu quirk mode. > > Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> > --- > drivers/gpu/drm/virtio/virtgpu_drv.h | 1 +

Since commit "9fdd90c0f4 drm/virtio: add virtio_gpu_alloc_fence()" fences are not allocated any more by virtio_gpu_fence_emit(). So there is no need to pass down a reference to the fence pointer, a plain pointer is enough now. Convert virtio_gpu_fence_emit() and callers. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_drv.h | 14

Move all remaining virtio_gpu_notify() calls from virtio_gpu_cmd_* to the callers, for consistency reasons. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_gem.c | 2 ++ drivers/gpu/drm/virtio/virtgpu_ioctl.c | 3 +++ drivers/gpu/drm/virtio/virtgpu_kms.c | 3 +++ drivers/gpu/drm/virtio/virtgpu_object.c | 1 + drivers/gpu/drm/virtio/virtgpu_vq.c

From: Xin He <hexin.op at bytedance.com> [ Upstream commit 836b194d65782aaec4485a07d2aab52d3f698505 ] Before setting shmem->pages to NULL, kfree() should be called. Signed-off-by: Xin He <hexin.op at bytedance.com> Reviewed-by: Qi Liu <liuqi.16 at bytedance.com> Link: http://patchwork.freedesktop.org/patch/msgid/20200722051851.72662-1-hexin.op at bytedance.com

From: Xin He <hexin.op at bytedance.com> [ Upstream commit 836b194d65782aaec4485a07d2aab52d3f698505 ] Before setting shmem->pages to NULL, kfree() should be called. Signed-off-by: Xin He <hexin.op at bytedance.com> Reviewed-by: Qi Liu <liuqi.16 at bytedance.com> Link: http://patchwork.freedesktop.org/patch/msgid/20200722051851.72662-1-hexin.op at bytedance.com

virtio-gpu basically needs a sg_table for the bo, to tell the host where the backing pages for the object are. So the gem shmem helpers are a perfect fit. Some drm_gem_object_funcs need thin wrappers to update the host state, but otherwise the helpers handle everything just fine. Once the fencing was sorted the switch was surprisingly easy and for the most part just removing the ttm code. v4:

virtio-gpu basically needs a sg_table for the bo, to tell the host where the backing pages for the object are. So the gem shmem helpers are a perfect fit. Some drm_gem_object_funcs need thin wrappers to update the host state, but otherwise the helpers handle everything just fine. Once the fencing was sorted the switch was surprisingly easy and for the most part just removing the ttm code.

virtio-gpu basically needs a sg_table for the bo, to tell the host where the backing pages for the object are. So the gem shmem helpers are a perfect fit. Some drm_gem_object_funcs need thin wrappers to update the host state, but otherwise the helpers handle everything just fine. Once the fencing was sorted the switch was surprisingly easy and for the most part just removing the ttm code. v5:

virtio-gpu basically needs a sg_table for the bo, to tell the host where the backing pages for the object are. So the gem shmem helpers are a perfect fit. Some drm_gem_object_funcs need thin wrappers to update the host state, but otherwise the helpers handle everything just fine. Once the fencing was sorted the switch was surprisingly easy and for the most part just removing the ttm code.

. On Wed, Jun 19, 2019 at 11:08 PM Gerd Hoffmann <kraxel at redhat.com> wrote: > > virtio-gpu basically needs a sg_table for the bo, to tell the host where > the backing pages for the object are. So the gem shmem helpers are a > perfect fit. Some drm_gem_object_funcs need thin wrappers to update the > host state, but otherwise the helpers handle everything just fine. >

On Wed, Feb 5, 2020 at 3:00 AM Gerd Hoffmann <kraxel at redhat.com> wrote: > > Introduce new virtio_gpu_object_shmem_init() helper function which will > create the virtio_gpu_mem_entry array, containing the backing storage > information for the host. For the most path this just moves code from > virtio_gpu_object_attach(). > > Signed-off-by: Gerd Hoffmann <kraxel at

Introduce new virtio_gpu_object_shmem_init() helper function which will create the virtio_gpu_mem_entry array, containing the backing storage information for the host. For the most path this just moves code from virtio_gpu_object_attach(). Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_drv.h | 4 +- drivers/gpu/drm/virtio/virtgpu_object.c | 55

Use gem reservation helpers and direct reservation_object_* calls instead of ttm. v3: Due to using the gem reservation object it is initialized and ready for use before calling ttm_bo_init, so we can also drop the tricky fence logic which checks whenever the command is in flight still. We can simply fence our object before submitting the virtio command and be done with it. Signed-off-by: Gerd

Rework fencing workflow. Stop using ttm helpers, use the virtio_gpu_array_* helpers instead. Due to using the gem reservation object it is initialized and ready for use before calling ttm_bo_init. So we can simply use the standard fencing workflow and drop the tricky logic which checks whenever the command is in flight still. v6: rewrite most of the patch. Signed-off-by: Gerd Hoffmann

On Tue, Sep 08, 2020 at 08:47:41AM +0200, Gerd Hoffmann wrote: > These days dma ops can be overridden per device, and the virtio core "can be overridden" or "are"? The comment above vring_use_dma_api() suggests that's not yet done. If that's wrong then I think updating the comment would be really good. -Daniel > uses that to handle the dma quirks transparently