On 2018-12-21 8:26 a.m., Emily Deng wrote:
> When the bo is used to set mode, the bo need to be pinned.
>
> Signed-off-by: Emily Deng <emily.d...@amd.com>
>
> [...]
>
> @@ -235,6 +222,47 @@ static bool dce_virtual_crtc_mode_fixup(struct drm_crtc
> *crtc,
> static int dce_virtual_crtc_set_base(struct drm_crtc *crtc, int x, int y,
> struct drm_framebuffer *old_fb)
> {
> + struct drm_framebuffer *target_fb;
> + struct drm_gem_object *obj;
> + struct amdgpu_bo *abo;
> + int r;
> +
> + /* no fb bound */
> + if (!crtc->primary->fb) {
> + DRM_DEBUG_KMS("No FB bound\n");
> + return 0;
> + }
> +
> + target_fb = crtc->primary->fb;
> +
> + obj = kcl_drm_fb_get_gem_obj(target_fb, 0);
> + abo = gem_to_amdgpu_bo(obj);
> + r = amdgpu_bo_reserve(abo, false);
> + if (unlikely(r != 0))
> + return r;
> +
> + r = amdgpu_bo_pin(abo, AMDGPU_GEM_DOMAIN_VRAM);
> + if (unlikely(r != 0)) {
> + amdgpu_bo_unreserve(abo);
> + return -EINVAL;
> + }
> +
> + amdgpu_bo_unreserve(abo);
> + return 0;
> +}
Good catch, but old_fb also needs to be unpinned, otherwise the pinning
is unbalanced and a BO will always be pinned after it's scanned out once.
> +static int dce_virtual_crtc_mode_set(struct drm_crtc *crtc,
> + struct drm_display_mode *mode,
> + struct drm_display_mode *adjusted_mode,
> + int x, int y, struct drm_framebuffer *old_fb)
Indentation of the continuation lines here looks wrong (needs 3 more
spaces if I'm counting correctly). Would be good to fix that up while
you're moving the function. :)
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
