linux-kernel - Re: [PATCH 3/4] drm: Add helper for simple display pipeline

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 9 May 2016 16:46:48 +0200
From:	Daniel Vetter <daniel@...ll.ch>
To:	Noralf Trønnes <noralf@...nnes.org>
Cc:	dri-devel@...ts.freedesktop.org, daniel@...ll.ch,
	treding@...dia.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/4] drm: Add helper for simple display pipeline

On Thu, May 05, 2016 at 03:24:33PM +0200, Noralf Trønnes wrote:
> Provides helper functions for drivers that have a simple display
> pipeline. Plane, crtc and encoder are collapsed into one entity.
> 
> Signed-off-by: Noralf Trønnes <noralf@...nnes.org>

> +static int drm_simple_kms_plane_atomic_check(struct drm_plane *plane,
> +					struct drm_plane_state *pstate)
> +{
> +	struct drm_simple_display_pipe *pipe;
> +	struct drm_crtc_state *cstate;
> +
> +	pipe = container_of(plane, struct drm_simple_display_pipe, plane);
> +	if (!pipe->funcs || !pipe->funcs->check)
> +		return 0;
> +
> +	cstate = drm_atomic_get_existing_crtc_state(pstate->state,
> +						    &pipe->crtc);
> +
> +	return pipe->funcs->check(pipe, pstate, cstate);
> +}

Ok one thing I've missed here is that for most drivers this is way too
simple a check function, which means we'll end up with tons of duplicated
code. Things which the drm core allows, but simple pipelines all don't
really cope with:
- plane scaling
- disabling the plane without the crtc (i.e. scan out black)
- plane not sized to fill the entire hactive/vactive

There's a helper to do most of these checks for you -
drm_plane_helper_check_update. I think it'd be good to place a call for
that in here, before we call down into the driver's ->check callback. But
ofc before we return 0; we want these checks always done. And catch all
these things so that drivers never fall over this pitfall.

Noticed while discussing tilcdc atomic patches, since tilcdc could
probably use drm_simple_display_pipe too.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch