Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Maxime Ripard <maxime@cerno.tech>
To:     Sam Ravnborg <sam@ravnborg.org>,
        Daniel Vetter <daniel.vetter@intel.com>,
        David Airlie <airlied@linux.ie>,
        Thierry Reding <thierry.reding@gmail.com>,
        Laurent Pinchart <Laurent.pinchart@ideasonboard.com>,
        Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
        Thomas Zimmermann <tzimmermann@suse.de>,
        Maxime Ripard <maxime@cerno.tech>,
        Jernej Skrabec <jernej.skrabec@gmail.com>,
        Neil Armstrong <narmstrong@baylibre.com>,
        Andrzej Hajda <a.hajda@samsung.com>,
        Jonas Karlman <jonas@kwiboo.se>,
        Robert Foss <robert.foss@linaro.org>
Cc:     linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org
Subject: [PATCH v2 0/8] drm/bridge: Make panel and bridge probe order consistent
Date:   Wed, 28 Jul 2021 15:32:21 +0200
Message-Id: <20210728133229.2247965-1-maxime@cerno.tech>
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi,=0D
=0D
We've encountered an issue with the RaspberryPi DSI panel that prevented th=
e=0D
whole display driver from probing.=0D
=0D
The issue is described in detail in the commit 7213246a803f ("drm/vc4: dsi:=
=0D
Only register our component once a DSI device is attached"), but the basic =
idea=0D
is that since the panel is probed through i2c, there's no synchronization=0D
between its probe and the registration of the MIPI-DSI host it's attached t=
o.=0D
=0D
We initially moved the component framework registration to the MIPI-DSI Hos=
t=0D
attach hook to make sure we register our component only when we have a DSI=
=0D
device attached to our MIPI-DSI host, and then use lookup our DSI device in=
 our=0D
bind hook.=0D
=0D
However, all the DSI bridges controlled through i2c are only registering th=
eir=0D
associated DSI device in their bridge attach hook, meaning with our change=
=0D
above, we never got that far, and therefore ended up in the same situation =
than=0D
the one we were trying to fix for panels.=0D
=0D
Since the RaspberryPi panel is the only driver in that situation, whereas i=
t=0D
seems like there's a consensus in bridge drivers, it makes more sense to tr=
y to=0D
mimic the bridge pattern in the panel driver.=0D
=0D
However, panels don't have an attach hook, and adding more panel hooks woul=
d=0D
lead to more path to maintain in each and every driver, while the general p=
ush=0D
is towards bridges. We also have to make sure that each and every DSI host =
and=0D
device driver behaves the same in order to have expectations to rely on.=0D
=0D
The solution I'm proposing is thus done in several steps:=0D
=0D
  - We get rid of the initial patch to make sure we support the bridge case=
,=0D
    and not the odd-panel one.=0D
=0D
  - Add a function that returns a bridge from a DT node, reducing the amoun=
t of=0D
    churn in each and every driver and making it a real incentive to not ca=
re=0D
    about panels in display drivers but only bridges.=0D
=0D
  - Add an attach and detach hook into the panel operations, and make it ca=
lled=0D
    automatically by the DRM panel bridge.=0D
=0D
  - Convert the VC4 DSI host to this new bridge function, and the Raspberry=
Pi=0D
    Panel to the new attach and detach hooks.=0D
=0D
If the general approach is agreed upon, other drivers will obviously be=0D
converted to drm_of_get_next.=0D
=0D
Let me know what you think,=0D
Maxime=0D
=0D
---=0D
=0D
Changes from v1:=0D
  - Change the name of drm_of_get_next function to drm_of_get_bridge=0D
  - Mention the revert of 87154ff86bf6 and squash the two patches that were=
=0D
    reverting that commit=0D
  - Add some documentation=0D
  - Make drm_panel_attach and _detach succeed when no callback is there=0D
=0D
Maxime Ripard (8):=0D
  Revert "drm/vc4: dsi: Only register our component once a DSI device is=0D
    attached"=0D
  drm/bridge: Add a function to abstract away panels=0D
  drm/bridge: Add documentation sections=0D
  drm/bridge: Document the probe issue with MIPI-DSI bridges=0D
  drm/panel: Create attach and detach callbacks=0D
  drm/vc4: dsi: Switch to drm_of_get_bridge=0D
  drm/panel: raspberrypi-touchscreen: Use the attach hook=0D
  drm/panel: raspberrypi-touchscreen: Remove MIPI-DSI driver=0D
=0D
 Documentation/gpu/drm-kms-helpers.rst         |  12 ++=0D
 drivers/gpu/drm/bridge/panel.c                |   6 +=0D
 drivers/gpu/drm/drm_bridge.c                  | 114 ++++++++++++-=0D
 drivers/gpu/drm/drm_of.c                      |   3 +=0D
 drivers/gpu/drm/drm_panel.c                   |  47 ++++++=0D
 .../drm/panel/panel-raspberrypi-touchscreen.c | 158 +++++++++---------=0D
 drivers/gpu/drm/vc4/vc4_drv.c                 |   2 +=0D
 drivers/gpu/drm/vc4/vc4_dsi.c                 |  53 +++---=0D
 include/drm/drm_bridge.h                      |   2 +=0D
 include/drm/drm_panel.h                       |  27 +++=0D
 10 files changed, 303 insertions(+), 121 deletions(-)=0D
=0D
-- =0D
2.31.1=0D
=0D