Re: [PATCH 4/4] drm/virtio: move virtio_gpu_mem_entry initialization to new function

From: Chia-I Wu
Date: Wed Feb 05 2020 - 14:57:17 EST


On Wed, Feb 5, 2020 at 3:00 AM Gerd Hoffmann <kraxel@xxxxxxxxxx> 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@xxxxxxxxxx>
> ---
> drivers/gpu/drm/virtio/virtgpu_drv.h | 4 ++
> drivers/gpu/drm/virtio/virtgpu_object.c | 49 +++++++++++++++++++++++++
> drivers/gpu/drm/virtio/virtgpu_vq.c | 49 ++-----------------------
> 3 files changed, 56 insertions(+), 46 deletions(-)
>
> diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
> index 15fb3c12f22f..be62a7469b04 100644
> --- a/drivers/gpu/drm/virtio/virtgpu_drv.h
> +++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
> @@ -71,6 +71,10 @@ struct virtio_gpu_object {
>
> struct sg_table *pages;
> uint32_t mapped;
> +
> + struct virtio_gpu_mem_entry *ents;
> + unsigned int nents;
> +
> bool dumb;
> bool created;
> };
> diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c
> index bce2b3d843fe..4e82e269a1f4 100644
> --- a/drivers/gpu/drm/virtio/virtgpu_object.c
> +++ b/drivers/gpu/drm/virtio/virtgpu_object.c
> @@ -121,6 +121,49 @@ struct drm_gem_object *virtio_gpu_create_object(struct drm_device *dev,
> return &bo->base.base;
> }
>
> +static int virtio_gpu_object_shmem_init(struct virtio_gpu_device *vgdev,
> + struct virtio_gpu_object *bo)
> +{
> + bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev);
> + struct scatterlist *sg;
> + int si, ret;
> +
> + ret = drm_gem_shmem_pin(&bo->base.base);
> + if (ret < 0)
> + return -EINVAL;
> +
> + bo->pages = drm_gem_shmem_get_sg_table(&bo->base.base);
> + if (bo->pages == NULL) {
> + drm_gem_shmem_unpin(&bo->base.base);
> + return -EINVAL;
> + }
> +
> + if (use_dma_api) {
> + bo->mapped = dma_map_sg(vgdev->vdev->dev.parent,
> + bo->pages->sgl, bo->pages->nents,
> + DMA_TO_DEVICE);
> + bo->nents = bo->mapped;
> + } else {
> + bo->nents = bo->pages->nents;
> + }
> +
> + bo->ents = kmalloc_array(bo->nents, sizeof(struct virtio_gpu_mem_entry),
> + GFP_KERNEL);
> + if (!bo->ents) {
> + DRM_ERROR("failed to allocate ent list\n");
> + return -ENOMEM;
> + }
> +
> + for_each_sg(bo->pages->sgl, sg, bo->nents, si) {
> + bo->ents[si].addr = cpu_to_le64(use_dma_api
> + ? sg_dma_address(sg)
> + : sg_phys(sg));
> + bo->ents[si].length = cpu_to_le32(sg->length);
> + bo->ents[si].padding = 0;
> + }
> + return 0;
> +}
> +
> int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
> struct virtio_gpu_object_params *params,
> struct virtio_gpu_object **bo_ptr,
> @@ -165,6 +208,12 @@ int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
> objs, fence);
> }
>
> + ret = virtio_gpu_object_shmem_init(vgdev, bo);
> + if (ret != 0) {
> + virtio_gpu_free_object(&shmem_obj->base);
> + return ret;
> + }
> +
> ret = virtio_gpu_object_attach(vgdev, bo, NULL);
> if (ret != 0) {
> virtio_gpu_free_object(&shmem_obj->base);
> diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
> index e258186bedb2..7db91376f2f2 100644
> --- a/drivers/gpu/drm/virtio/virtgpu_vq.c
> +++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
> @@ -1081,54 +1081,11 @@ int virtio_gpu_object_attach(struct virtio_gpu_device *vgdev,
> struct virtio_gpu_object *obj,
> struct virtio_gpu_fence *fence)
> {
> - bool use_dma_api = !virtio_has_iommu_quirk(vgdev->vdev);
> - struct virtio_gpu_mem_entry *ents;
> - struct scatterlist *sg;
> - int si, nents, ret;
> -
> - if (WARN_ON_ONCE(!obj->created))
> - return -EINVAL;
> - if (WARN_ON_ONCE(obj->pages))
> - return -EINVAL;
> -
> - ret = drm_gem_shmem_pin(&obj->base.base);
> - if (ret < 0)
> - return -EINVAL;
> -
> - obj->pages = drm_gem_shmem_get_sg_table(&obj->base.base);
> - if (obj->pages == NULL) {
> - drm_gem_shmem_unpin(&obj->base.base);
> - return -EINVAL;
> - }
> -
> - if (use_dma_api) {
> - obj->mapped = dma_map_sg(vgdev->vdev->dev.parent,
> - obj->pages->sgl, obj->pages->nents,
> - DMA_TO_DEVICE);
> - nents = obj->mapped;
> - } else {
> - nents = obj->pages->nents;
> - }
> -
> - /* gets freed when the ring has consumed it */
> - ents = kmalloc_array(nents, sizeof(struct virtio_gpu_mem_entry),
> - GFP_KERNEL);
> - if (!ents) {
> - DRM_ERROR("failed to allocate ent list\n");
> - return -ENOMEM;
> - }
> -
> - for_each_sg(obj->pages->sgl, sg, nents, si) {
> - ents[si].addr = cpu_to_le64(use_dma_api
> - ? sg_dma_address(sg)
> - : sg_phys(sg));
> - ents[si].length = cpu_to_le32(sg->length);
> - ents[si].padding = 0;
> - }
> -
> virtio_gpu_cmd_resource_attach_backing(vgdev, obj->hw_res_handle,
> - ents, nents,
> + obj->ents, obj->nents,
> fence);
> + obj->ents = NULL;
> + obj->nents = 0;
Hm, if the entries are temporary, can we allocate and initialize them
in this function?

virtio_gpu_object_shmem_init will just pin and map pages. Maybe
rename it to virtio_gpu_object_pin_pages (and add a helper
virtio_gpu_object_unpin_pages for use by virtio_gpu_cleanup_object).

Because we pin pages on object creation, virtio_gpu_gem_funcs does not
need to provide the optional pin/unpin hooks.

> return 0;
> }
>
> --
> 2.18.1
>