mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Romain TISSERAND <romain.tisserand@gmail.com>
To: Chen-Yu Tsai <wens@kernel.org>,
	Maxime Ripard <mripard@kernel.org>,
	Jernej Skrabec <jernej.skrabec@gmail.com>,
	Samuel Holland <samuel@sholland.org>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	Thomas Zimmermann <tzimmermann@suse.de>,
	David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
	dri-devel@lists.freedesktop.org, linux-sunxi@lists.linux.dev,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, sashiko-reviews@lists.linux.dev
Subject: [PATCH] drm/sun4i: dsi: set the connector orientation from the panel node
Date: Sun,  1 Nov 2026 21:34:06 +0100	[thread overview]
Message-ID: <20261101203406.1360656-1-romain.tisserand@gmail.com> (raw)

sun6i_dsi_attach() only accepts the panel once the DRM device is
registered, and returns -EPROBE_DEFER until then. By that time the
connector can no longer get a "panel orientation" property: it has to be
attached before registration. The fbdev emulation is also set up from the
hotplug event of the attach, so it gets no rotation hint either.

As a result the rotation property of a DSI panel on these SoCs is ignored:
on the PocketGo S30 (A33, 320x480 panel mounted for landscape use,
rotation = <270>) the console and the boot splash stay in portrait.

The panel is a child node of the DSI host, so read its orientation from
the device tree when binding, before the device is registered, and set it
on the connector then.

Signed-off-by: Romain TISSERAND <romain.tisserand@gmail.com>
---
Notes:
- Testing: built with W=1 for arm (multi_v7_defconfig, DRM_SUN6I_DSI=y
  and =m) and arm64 (defconfig, =m), no warnings. Tested on a PocketGo
  S30 (A33, ILI9488 DSI panel, rotation = <270>): on 6.18.54 with the same
  change (6.18 still calls the helper of_drm_get_panel_orientation()) the
  console and plymouth come up in landscape, where they were in portrait;
  on 7.3.1 with this patch the DSI connector reports "panel orientation"
  = "Left Side Up".
  Not tested on other boards; boards without a rotation property keep the
  current behaviour (no property is attached).
- Tool disclosure: this patch was prepared with Claude Opus 5.5
  (Anthropic) as a coding assistant, in an interactive session where I
  asked it to fix the console orientation on the PocketGo S30. It traced
  why the connector never got the orientation, wrote the change and the
  message. I reviewed and understand it, and tested it as described.

 drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c | 28 ++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
index d504ae5832..1eb8a4cde0 100644
--- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
+++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c
@@ -22,6 +22,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_mipi_dsi.h>
+#include <drm/drm_of.h>
 #include <drm/drm_panel.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
@@ -1055,6 +1056,31 @@ static const struct regmap_config sun6i_dsi_regmap_config = {
 	.name		= "mipi-dsi",
 };
 
+/*
+ * The panel only attaches once the DRM device is registered, too late to
+ * give the connector its orientation (the property cannot be added after
+ * registration), so fbcon and the boot splash would not rotate. Read the
+ * orientation from the panel node, a child of the DSI node, at bind time.
+ */
+static void sun6i_dsi_set_panel_orientation(struct sun6i_dsi *dsi)
+{
+	enum drm_panel_orientation orientation;
+	struct device_node *np;
+
+	for_each_available_child_of_node(dsi->dev->of_node, np) {
+		if (of_node_name_eq(np, "port") || of_node_name_eq(np, "ports"))
+			continue;
+
+		if (!drm_of_get_panel_orientation(np, &orientation) &&
+		    orientation != DRM_MODE_PANEL_ORIENTATION_UNKNOWN) {
+			drm_connector_set_panel_orientation(&dsi->connector,
+							    orientation);
+			of_node_put(np);
+			return;
+		}
+	}
+}
+
 static int sun6i_dsi_bind(struct device *dev, struct device *master,
 			 void *data)
 {
@@ -1085,6 +1111,8 @@ static int sun6i_dsi_bind(struct device *dev, struct device *master,
 
 	drm_connector_attach_encoder(&dsi->connector, &dsi->encoder);
 
+	sun6i_dsi_set_panel_orientation(dsi);
+
 	dsi->drm = drm;
 
 	return 0;

base-commit: a1e63ab695ad76850ab9787b9eb6aab695dc2965
-- 
2.43.0


                 reply	other threads:[~2026-11-01 20:34 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20261101203406.1360656-1-romain.tisserand@gmail.com \
    --to=romain.tisserand@gmail.com \
    --cc=airlied@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=jernej.skrabec@gmail.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-sunxi@lists.linux.dev \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=mripard@kernel.org \
    --cc=samuel@sholland.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=simona@ffwll.ch \
    --cc=tzimmermann@suse.de \
    --cc=wens@kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®