All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
To: Maxime Ripard <maxime@cerno.tech>
Cc: David Airlie <airlied@linux.ie>,
	Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	DRI Development <dri-devel@lists.freedesktop.org>,
	Eric Anholt <eric@anholt.net>,
	bcm-kernel-feedback-list@broadcom.com,
	linux-rpi-kernel@lists.infradead.org,
	Thomas Zimmermann <tzimmermann@suse.de>,
	Daniel Vetter <daniel.vetter@intel.com>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v6 5/9] drm/vc4: hdmi: Create a custom connector state
Date: Tue, 15 Dec 2020 11:02:28 +0000	[thread overview]
Message-ID: <CAPY8ntAo0VOHRBh32-k5+v_E0E91pbhaC-An0Hhv+_nWFSkyYQ@mail.gmail.com> (raw)
In-Reply-To: <20201210142329.10640-6-maxime@cerno.tech>

Hi Maxime

On Thu, 10 Dec 2020 at 14:23, Maxime Ripard <maxime@cerno.tech> wrote:
>
> When run with a higher bpc than 8, the clock of the HDMI controller needs
> to be adjusted. Let's create a connector state that will be used at
> atomic_check and atomic_enable to compute and store the clock rate
> associated to the state.
>
> Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
> Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
> Signed-off-by: Maxime Ripard <maxime@cerno.tech>
> ---
>  drivers/gpu/drm/vc4/vc4_hdmi.c | 33 ++++++++++++++++++++++++++++++---
>  drivers/gpu/drm/vc4/vc4_hdmi.h | 10 ++++++++++
>  2 files changed, 40 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
> index 61039cc89d9d..8978df3f0ca4 100644
> --- a/drivers/gpu/drm/vc4/vc4_hdmi.c
> +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
> @@ -170,10 +170,37 @@ static int vc4_hdmi_connector_get_modes(struct drm_connector *connector)
>
>  static void vc4_hdmi_connector_reset(struct drm_connector *connector)
>  {
> -       drm_atomic_helper_connector_reset(connector);
> +       struct vc4_hdmi_connector_state *old_state =
> +               conn_state_to_vc4_hdmi_conn_state(connector->state);
> +       struct vc4_hdmi_connector_state *new_state =
> +               kzalloc(sizeof(*conn_state), GFP_KERNEL);
>
>         if (connector->state)
> -               drm_atomic_helper_connector_tv_reset(connector);
> +               __drm_atomic_helper_connector_destroy_state(connector->state);
> +
> +       kfree(old_state);
> +
> +       if (!new_state)
> +               return;

This has changed since v5 that I added my R-B to.

So we no longer call __drm_atomic_helper_connector_reset should the
kzalloc fail. Doesn't that mean connector->state is still set to
old_state? Except we've called kfree on that and it's therefore
invalid.

Calling __drm_atomic_helper_connector_reset unconditionally is fine as
it checks for the new pointer being NULL before dereferencing, but
always assigns it to connector->state.
Calling drm_atomic_helper_connector_tv_reset when connector->state =
NULL isn't safe.

> +       __drm_atomic_helper_connector_reset(connector, &new_state->base);

Put the if (!new_state) return; here?
Patch 9/9 can set the max_bpc and max_bpc_requested field here too.

  Dave

> +       drm_atomic_helper_connector_tv_reset(connector);
> +}
> +
> +static struct drm_connector_state *
> +vc4_hdmi_connector_duplicate_state(struct drm_connector *connector)
> +{
> +       struct drm_connector_state *conn_state = connector->state;
> +       struct vc4_hdmi_connector_state *vc4_state = conn_state_to_vc4_hdmi_conn_state(conn_state);
> +       struct vc4_hdmi_connector_state *new_state;
> +
> +       new_state = kzalloc(sizeof(*new_state), GFP_KERNEL);
> +       if (!new_state)
> +               return NULL;
> +
> +       __drm_atomic_helper_connector_duplicate_state(connector, &new_state->base);
> +
> +       return &new_state->base;
>  }
>
>  static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
> @@ -181,7 +208,7 @@ static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
>         .fill_modes = drm_helper_probe_single_connector_modes,
>         .destroy = vc4_hdmi_connector_destroy,
>         .reset = vc4_hdmi_connector_reset,
> -       .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
> +       .atomic_duplicate_state = vc4_hdmi_connector_duplicate_state,
>         .atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
>  };
>
> diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
> index 0526a9cf608a..2cf5362052e2 100644
> --- a/drivers/gpu/drm/vc4/vc4_hdmi.h
> +++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
> @@ -180,6 +180,16 @@ encoder_to_vc4_hdmi(struct drm_encoder *encoder)
>         return container_of(_encoder, struct vc4_hdmi, encoder);
>  }
>
> +struct vc4_hdmi_connector_state {
> +       struct drm_connector_state      base;
> +};
> +
> +static inline struct vc4_hdmi_connector_state *
> +conn_state_to_vc4_hdmi_conn_state(struct drm_connector_state *conn_state)
> +{
> +       return container_of(conn_state, struct vc4_hdmi_connector_state, base);
> +}
> +
>  void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
>                        struct drm_display_mode *mode);
>  void vc4_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi);
> --
> 2.28.0
>

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

WARNING: multiple messages have this Message-ID (diff)
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
To: Maxime Ripard <maxime@cerno.tech>
Cc: David Airlie <airlied@linux.ie>,
	DRI Development <dri-devel@lists.freedesktop.org>,
	bcm-kernel-feedback-list@broadcom.com,
	linux-rpi-kernel@lists.infradead.org,
	Thomas Zimmermann <tzimmermann@suse.de>,
	Daniel Vetter <daniel.vetter@intel.com>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v6 5/9] drm/vc4: hdmi: Create a custom connector state
Date: Tue, 15 Dec 2020 11:02:28 +0000	[thread overview]
Message-ID: <CAPY8ntAo0VOHRBh32-k5+v_E0E91pbhaC-An0Hhv+_nWFSkyYQ@mail.gmail.com> (raw)
In-Reply-To: <20201210142329.10640-6-maxime@cerno.tech>

Hi Maxime

On Thu, 10 Dec 2020 at 14:23, Maxime Ripard <maxime@cerno.tech> wrote:
>
> When run with a higher bpc than 8, the clock of the HDMI controller needs
> to be adjusted. Let's create a connector state that will be used at
> atomic_check and atomic_enable to compute and store the clock rate
> associated to the state.
>
> Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
> Reviewed-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
> Signed-off-by: Maxime Ripard <maxime@cerno.tech>
> ---
>  drivers/gpu/drm/vc4/vc4_hdmi.c | 33 ++++++++++++++++++++++++++++++---
>  drivers/gpu/drm/vc4/vc4_hdmi.h | 10 ++++++++++
>  2 files changed, 40 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
> index 61039cc89d9d..8978df3f0ca4 100644
> --- a/drivers/gpu/drm/vc4/vc4_hdmi.c
> +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
> @@ -170,10 +170,37 @@ static int vc4_hdmi_connector_get_modes(struct drm_connector *connector)
>
>  static void vc4_hdmi_connector_reset(struct drm_connector *connector)
>  {
> -       drm_atomic_helper_connector_reset(connector);
> +       struct vc4_hdmi_connector_state *old_state =
> +               conn_state_to_vc4_hdmi_conn_state(connector->state);
> +       struct vc4_hdmi_connector_state *new_state =
> +               kzalloc(sizeof(*conn_state), GFP_KERNEL);
>
>         if (connector->state)
> -               drm_atomic_helper_connector_tv_reset(connector);
> +               __drm_atomic_helper_connector_destroy_state(connector->state);
> +
> +       kfree(old_state);
> +
> +       if (!new_state)
> +               return;

This has changed since v5 that I added my R-B to.

So we no longer call __drm_atomic_helper_connector_reset should the
kzalloc fail. Doesn't that mean connector->state is still set to
old_state? Except we've called kfree on that and it's therefore
invalid.

Calling __drm_atomic_helper_connector_reset unconditionally is fine as
it checks for the new pointer being NULL before dereferencing, but
always assigns it to connector->state.
Calling drm_atomic_helper_connector_tv_reset when connector->state =
NULL isn't safe.

> +       __drm_atomic_helper_connector_reset(connector, &new_state->base);

Put the if (!new_state) return; here?
Patch 9/9 can set the max_bpc and max_bpc_requested field here too.

  Dave

> +       drm_atomic_helper_connector_tv_reset(connector);
> +}
> +
> +static struct drm_connector_state *
> +vc4_hdmi_connector_duplicate_state(struct drm_connector *connector)
> +{
> +       struct drm_connector_state *conn_state = connector->state;
> +       struct vc4_hdmi_connector_state *vc4_state = conn_state_to_vc4_hdmi_conn_state(conn_state);
> +       struct vc4_hdmi_connector_state *new_state;
> +
> +       new_state = kzalloc(sizeof(*new_state), GFP_KERNEL);
> +       if (!new_state)
> +               return NULL;
> +
> +       __drm_atomic_helper_connector_duplicate_state(connector, &new_state->base);
> +
> +       return &new_state->base;
>  }
>
>  static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
> @@ -181,7 +208,7 @@ static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
>         .fill_modes = drm_helper_probe_single_connector_modes,
>         .destroy = vc4_hdmi_connector_destroy,
>         .reset = vc4_hdmi_connector_reset,
> -       .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
> +       .atomic_duplicate_state = vc4_hdmi_connector_duplicate_state,
>         .atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
>  };
>
> diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
> index 0526a9cf608a..2cf5362052e2 100644
> --- a/drivers/gpu/drm/vc4/vc4_hdmi.h
> +++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
> @@ -180,6 +180,16 @@ encoder_to_vc4_hdmi(struct drm_encoder *encoder)
>         return container_of(_encoder, struct vc4_hdmi, encoder);
>  }
>
> +struct vc4_hdmi_connector_state {
> +       struct drm_connector_state      base;
> +};
> +
> +static inline struct vc4_hdmi_connector_state *
> +conn_state_to_vc4_hdmi_conn_state(struct drm_connector_state *conn_state)
> +{
> +       return container_of(conn_state, struct vc4_hdmi_connector_state, base);
> +}
> +
>  void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
>                        struct drm_display_mode *mode);
>  void vc4_hdmi_phy_disable(struct vc4_hdmi *vc4_hdmi);
> --
> 2.28.0
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  parent reply	other threads:[~2020-12-15 11:04 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-10 14:23 [PATCH v6 0/9] drm/vc4: hdmi: Support the 10/12 bit output Maxime Ripard
2020-12-10 14:23 ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 1/9] drm/vc4: hvs: Align the HVS atomic hooks to the new API Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 2/9] drm/vc4: Pass the atomic state to encoder hooks Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 3/9] drm/vc4: hdmi: Take into account the clock doubling flag in atomic_check Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 4/9] drm/vc4: hdmi: Don't access the connector state in reset if kmalloc fails Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 5/9] drm/vc4: hdmi: Create a custom connector state Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:26   ` Maxime Ripard
2020-12-10 14:26     ` Maxime Ripard
2020-12-15 11:02   ` Dave Stevenson [this message]
2020-12-15 11:02     ` Dave Stevenson
2020-12-10 14:23 ` [PATCH v6 6/9] drm/vc4: hdmi: Store pixel frequency in the " Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 7/9] drm/vc4: hdmi: Use the connector state pixel rate for the PHY Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 8/9] drm/vc4: hdmi: Limit the BCM2711 to the max without scrambling Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-10 14:23 ` [PATCH v6 9/9] drm/vc4: hdmi: Enable 10/12 bpc output Maxime Ripard
2020-12-10 14:23   ` Maxime Ripard
2020-12-15 11:04   ` Dave Stevenson
2020-12-15 11:04     ` Dave Stevenson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAPY8ntAo0VOHRBh32-k5+v_E0E91pbhaC-An0Hhv+_nWFSkyYQ@mail.gmail.com \
    --to=dave.stevenson@raspberrypi.com \
    --cc=airlied@linux.ie \
    --cc=bcm-kernel-feedback-list@broadcom.com \
    --cc=daniel.vetter@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=eric@anholt.net \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-rpi-kernel@lists.infradead.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=maxime@cerno.tech \
    --cc=tzimmermann@suse.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.