[PATCH i915 v2 1/2] i915: wait for fences in mmio_flip()

[Chris Wilson]

On Fri, Oct 30, 2015 at 06:03:50PM -0700, Alex Goins wrote:
> If a buffer is backed by dmabuf, wait on its reservation object's fences
> before flipping.
> 
> Signed-off-by: Alex Goins <agoins at nvidia.com>
> ---
>  drivers/gpu/drm/i915/intel_display.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index b2270d5..1485640 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -44,6 +44,8 @@
>  #include <drm/drm_plane_helper.h>
>  #include <drm/drm_rect.h>
>  #include <linux/dma_remapping.h>
> +#include <linux/reservation.h>
> +#include <linux/dma-buf.h>
>  
>  /* Primary plane formats for gen <= 3 */
>  static const uint32_t i8xx_primary_formats[] = {
> @@ -11170,10 +11172,21 @@ static void ilk_do_mmio_flip(struct intel_crtc *intel_crtc)
>  static void intel_do_mmio_flip(struct intel_crtc *intel_crtc)
>  {
>  	struct drm_device *dev = intel_crtc->base.dev;
> +	struct drm_i915_gem_object *pending_flip_obj =
> +		intel_crtc->unpin_work->pending_flip_obj;
>  	u32 start_vbl_count;
>  
>  	intel_mark_page_flip_active(intel_crtc);
>  
> +	/* For framebuffer backed by dmabuf, wait for fence */
> +	mutex_lock(&dev->object_name_lock);

The lock here is unfortunate. I thought once a dmabuf as attached to an
object, it persists until the object is destroyed, so afaict the lock
here is unnecessary (as it only protects against a userspace race in
attaching a dmabuf).

> +	if (pending_flip_obj->base.dma_buf) {
> +		reservation_object_wait_timeout_rcu(

Side-question, are these fences exclusive or do we track read/write?
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre