Re: [PATCH v4 02/12] drm/client: Restrict the plane_state scope

From: Jani Nikula <jani.nikula@linux.intel.com>
To: Maxime Ripard <maxime.ripard@bootlin.com>
Cc: eben@raspberrypi.org, David Airlie <airlied@linux.ie>,
	Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	dri-devel@lists.freedesktop.org,
	Paul Kocialkowski <paul.kocialkowski@bootlin.com>,
	Sean Paul <seanpaul@chromium.org>,
	Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
	Daniel Vetter <daniel.vetter@intel.com>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v4 02/12] drm/client: Restrict the plane_state scope
Date: Mon, 17 Jun 2019 12:27:37 +0300	[thread overview]
Message-ID: <87a7egv9x2.fsf@intel.com> (raw)
In-Reply-To: <20190614141211.rl7ihqgzllcai634@flea>

On Fri, 14 Jun 2019, Maxime Ripard <maxime.ripard@bootlin.com> wrote:
> Hi Jani,
>
> On Fri, Jun 14, 2019 at 03:28:59PM +0300, Jani Nikula wrote:
>> On Fri, 14 Jun 2019, Maxime Ripard <maxime.ripard@bootlin.com> wrote:
>> > The drm_client_modeset_commit_atomic function uses two times the
>> > plane_state variable in inner blocks of code, but the variable has a scope
>> > global to this function.
>> >
>> > This will lead to inadvertent devs to reuse the variable in the second
>> > block with the value left by the first, without any warning from the
>> > compiler since value would have been initialized.
>> >
>> > Fix this by moving the variable declaration to the proper scope.
>>
>> This is an improvement, but I'd consider renaming also to not shadow
>> variables.
>>
>> > Signed-off-by: Maxime Ripard <maxime.ripard@bootlin.com>
>> > ---
>> >  drivers/gpu/drm/drm_client_modeset.c | 5 ++++-
>> >  1 file changed, 4 insertions(+), 1 deletion(-)
>> >
>> > diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c
>> > index 006bf7390e7d..8264c3a732b0 100644
>> > --- a/drivers/gpu/drm/drm_client_modeset.c
>> > +++ b/drivers/gpu/drm/drm_client_modeset.c
>> > @@ -861,7 +861,6 @@ EXPORT_SYMBOL(drm_client_panel_rotation);
>> >  static int drm_client_modeset_commit_atomic(struct drm_client_dev *client, bool active)
>> >  {
>> >  	struct drm_device *dev = client->dev;
>> > -	struct drm_plane_state *plane_state;
>> >  	struct drm_plane *plane;
>> >  	struct drm_atomic_state *state;
>> >  	struct drm_modeset_acquire_ctx ctx;
>> > @@ -879,6 +878,8 @@ static int drm_client_modeset_commit_atomic(struct drm_client_dev *client, bool
>> >  	state->acquire_ctx = &ctx;
>> >  retry:
>> >  	drm_for_each_plane(plane, dev) {
>> > +		struct drm_plane_state *plane_state;
>> > +
>> >  		plane_state = drm_atomic_get_plane_state(state, plane);
>> >  		if (IS_ERR(plane_state)) {
>> >  			ret = PTR_ERR(plane_state);
>> > @@ -901,6 +902,8 @@ static int drm_client_modeset_commit_atomic(struct drm_client_dev *client, bool
>> >  		unsigned int rotation;
>> >
>> >  		if (drm_client_panel_rotation(mode_set, &rotation)) {
>> > +			struct drm_plane_state *plane_state;
>> > +
>
> That's not super clear from that patch, but this variable will not
> shadow the first one.
>
> The code layout is pretty much this one:
>
> loop () {
>   struct drm_plane_state *plane_state;
>
>   [...]
> }
>
> loop () {
>   loop () {
>     struct drm_plane_state *plane_state;
>
>     [...]
>   }
> }
>
> so the shadowing doesn't exist

Indeed, sorry for the noise.

BR,
Jani.

-- 
Jani Nikula, Intel Open Source Graphics Center

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel