On Mon, Jun 17, 2019 at 01:14:02PM +0200, Gerd Hoffmann wrote: > ttm increasingly gets into the way while hacking on virtio-gpu memory > management. So maybe we are better off without, using gem helpers > instead? > > Trying remove the dependency on ttm reservations looks like a good first > first step to me. This patch series is a (small) start on that. > > At the point I'm looking for comments whevener the plan and the initial > patches look sane and for things I should possibly watch out for on the > way. This matches my own plan for aligning gem and ttm more, so +1 from me. Long-term I'm not sure whether we should embed a drm_gem_object into the ttm_buffer_object, or whether we should extract more helpers from ttm into gem helpers, or something in-between. But there's a bunch of lower-hanging fruit to take care of first. Cheers, Daniel > > thanks, > Gerd > > Gerd Hoffmann (4): > drm/virtio: pass gem reservation object to ttm init > drm/virtio: switch virtio_gpu_wait_ioctl() to gem helper. > drm/virtio: simplify cursor updates > drm/virtio: remove virtio_gpu_object_wait > > drivers/gpu/drm/virtio/virtgpu_drv.h | 1 - > drivers/gpu/drm/virtio/virtgpu_ioctl.c | 18 ++++-------------- > drivers/gpu/drm/virtio/virtgpu_object.c | 16 ++-------------- > drivers/gpu/drm/virtio/virtgpu_plane.c | 13 +++---------- > 4 files changed, 9 insertions(+), 39 deletions(-) > > -- > 2.18.1 > > _______________________________________________ > dri-devel mailing list > dri-devel@xxxxxxxxxxxxxxxxxxxxx > https://lists.freedesktop.org/mailman/listinfo/dri-devel -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel