[Intel-gfx] [PATCH 1/2] drm/atomic-helper: Complete fake_commit->flip_done potentially earlier

[Maarten Lankhorst]

Op 22-11-18 om 15:34 schreef Ville Syrjala:
> From: Ville Syrjälä <ville.syrjala at linux.intel.com>
>
> Consider the following scenario:
> 1. nonblocking enable crtc
> 2. wait for the event
> 3. nonblocking disable crtc
>
> On i915 this can lead to a spurious -EBUSY from step 3 on
> account of non-enabled planes getting the fake_commit in step 1
> and we don't complete the fake_commit-> flip_done until
> drm_atomic_helper_commit_hw_done() which can happen a long
> time after the flip event was sent out.
>
> This will become somewhat easy to hit on SKL+ once we start
> to add all the planes for the crtc to every modeset commit
> for the purposes of forcing a watermark register programming
> [1].
>
> To make the race a little less pronounced let's complete
> fake_commit->flip_done after drm_atomic_helper_wait_for_flip_done().
> For the single crtc case this should make the race quite
> theoretical, assuming drm_atomic_helper_wait_for_flip_done()
> actually has to wait for the real commit flip_done. In case
> the real commit flip_done gets completed singificantly before
> drm_atomic_helper_wait_for_flip_done(), or we are dealing with
> multiple crtcs whose vblanks don't line up nicely the race still
> exists.
>
> [1] https://patchwork.freedesktop.org/patch/262670/
>
> Cc: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
> Fixes: 080de2e5be2d ("drm/atomic: Check for busy planes/connectors before setting the commit")
> Testcase: igt/kms_cursor_legacy/*nonblocking-modeset-vs-cursor-atomic
> Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> ---
>  drivers/gpu/drm/drm_atomic_helper.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
> index fa95f9974f6d..47398662b895 100644
> --- a/drivers/gpu/drm/drm_atomic_helper.c
> +++ b/drivers/gpu/drm/drm_atomic_helper.c
> @@ -1460,6 +1460,9 @@ void drm_atomic_helper_wait_for_flip_done(struct drm_device *dev,
>  			DRM_ERROR("[CRTC:%d:%s] flip_done timed out\n",
>  				  crtc->base.id, crtc->name);
>  	}
> +
> +	if (old_state->fake_commit)
> +		complete_all(&old_state->fake_commit->flip_done);
>  }
>  EXPORT_SYMBOL(drm_atomic_helper_wait_for_flip_done);
>  

Yeah as long as we don't enable hw_done sooner, this should be fine.

For both patches:

Reviewed-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>