Re: [PATCH] drm/vmwgfx: Restore Guest-Backed only cursor plane support

[Zack Rusin]

On Wed, Oct 29, 2025 at 10:16 AM Ian Forbes <ian.for...@broadcom.com> wrote:
>
> The referenced fixes commit broke the cursor plane for configurations
> which have Guest-Backed surfaces but no cursor MOB support.
>
> Fixes: 965544150d1c ("drm/vmwgfx: Refactor cursor handling")
> Signed-off-by: Ian Forbes <ian.for...@broadcom.com>
> ---
>  drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c | 14 ++++++++++++++
>  drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.h |  1 +
>  2 files changed, 15 insertions(+)
>
> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c 
> b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c
> index 718832b08d96..87f0f61e470c 100644
> --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c
> +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c
> @@ -100,6 +100,8 @@ vmw_cursor_update_type(struct vmw_private *vmw, struct 
> vmw_plane_state *vps)
>         if (vmw->has_mob) {
>                 if ((vmw->capabilities2 & SVGA_CAP2_CURSOR_MOB) != 0)
>                         return VMW_CURSOR_UPDATE_MOB;
> +               else
> +                       return VMW_CURSOR_UPDATE_GB_ONLY;
>         }
>
>         return VMW_CURSOR_UPDATE_NONE;

Could you add a drm_warn_once before this return to give us a clear
clue what's happening here in the future?

Otherwise it looks great, thanks for fixing this!

z

smime.p7s
Description: S/MIME Cryptographic Signature