[Intel-gfx] [PATCH] drm/i915: sprite init failure on pre-SNB is not a failure

[Daniel Vetter]

On Thu, Jan 12, 2012 at 17:08, Jesse Barnes <jbarnes at virtuousgeek.org> wrote:
> We can call the plane init function unconditionally, but don't need to
> complain if it fails, since that will only happen if we're out of
> memory (and other things will fail) or if we're on the wrong platform
> (which is ok).
>
> And remove the DRM_ERRORs from the sprite code itself to avoid dmesg
> spam.
>
> Signed-off-by: Jesse Barnes <jbarnes at virtuousgeek.org>
>
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 29743de..e3bfa50 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -9027,12 +9027,9 @@ void intel_modeset_init(struct drm_device *dev)
>
>        for (i = 0; i < dev_priv->num_pipe; i++) {
>                intel_crtc_init(dev, i);
> -               if (HAS_PCH_SPLIT(dev)) {
> -                       ret = intel_plane_init(dev, i);
> -                       if (ret)
> -                               DRM_ERROR("plane %d init failed: %d\n",
> -                                         i, ret);
> -               }
> +               ret = intel_plane_init(dev, i);
> +               if (ret)
> +                       DRM_DEBUG_KMS("plane %d init failed: %d\n", i, ret);
>        }
>
>        /* Just disable it once at startup */
> diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
> index b26e7c4..32909e3 100644
> --- a/drivers/gpu/drm/i915/intel_sprite.c
> +++ b/drivers/gpu/drm/i915/intel_sprite.c
> @@ -466,10 +466,8 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
>        mutex_lock(&dev->struct_mutex);
>
>        ret = intel_pin_and_fence_fb_obj(dev, obj, NULL);
> -       if (ret) {
> -               DRM_ERROR("failed to pin object\n");
> +       if (ret)
>                goto out_unlock;
> -       }
>
>        intel_plane->obj = obj;
>
> @@ -632,10 +630,8 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe)
>        unsigned long possible_crtcs;
>        int ret;
>
> -       if (!(IS_GEN6(dev) || IS_GEN7(dev))) {
> -               DRM_ERROR("new plane code only for SNB+\n");
> +       if (!(IS_GEN6(dev) || IS_GEN7(dev)))
>                return -ENODEV;
> -       }

Why not return 0; to avoid the "plane init" spam at kms debug level?
After all, nothing failed, there's just no plane there to init ...
-Daniel
-- 
Daniel Vetter
daniel.vetter at ffwll.ch - +41 (0) 79 364 57 48 - http://blog.ffwll.ch