Skip to contentSkip to conversation

drm: bridge: anx7688 and mux drivers

PATCH RESEND In review Hsin-Yi Wang 6 participants 25 messages Dec 11, 2019 – Dec 20, 2019

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

cover drm: bridge: anx7688 and mux drivers

This is a resend of [1] with a few modification due to drm core function changes and use regmap abstraction.

The gpio mux driver is required for MT8173 board layout:

                                  /-- anx7688

-- MT8173 HDMI bridge -- GPIO mux
\-- native HDMI

[1] https://lore.kernel.org/lkml/[email protected]/

Nicolas Boichat (4):
dt-bindings: drm/bridge: analogix-anx7688: Add ANX7688 transmitter
binding
drm: bridge: anx7688: Add anx7688 bridge driver support.
dt-bindings: drm/bridge: Add GPIO display mux binding
drm: bridge: Generic GPIO mux driver

 .../bindings/display/bridge/anx7688.yaml      |  60 ++++
 .../bindings/display/bridge/gpio-mux.yaml     |  89 +++++
 drivers/gpu/drm/bridge/Kconfig                |  19 ++
 drivers/gpu/drm/bridge/Makefile               |   2 +
 drivers/gpu/drm/bridge/analogix-anx7688.c     | 202 ++++++++++++
 drivers/gpu/drm/bridge/generic-gpio-mux.c     | 306 ++++++++++++++++++
 6 files changed, 678 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/anx7688.yaml
 create mode 100644 Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
 create mode 100644 drivers/gpu/drm/bridge/generic-gpio-mux.c
Signature · 1 line
2.24.0.525.g8f36a354ae-goog
#1 lore raw

PATCH RESEND 1/4 dt-bindings: drm/bridge: analogix-anx7688: Add ANX7688 transmitter binding

From: Nicolas Boichat <[email protected]>

Add support for analogix,anx7688

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/anx7688.yaml      | 60 +++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/anx7688.yaml
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +60 -0
new file mode 100644
index 000000000000..cf79f7cf8fdf
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/anx7688.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Analogix ANX7688 SlimPort (Single-Chip Transmitter for DP over USB-C)
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  The ANX7688 is a single-chip mobile transmitter to support 4K 60 frames per
+  second (4096x2160p60) or FHD 120 frames per second (1920x1080p120) video
+  resolution from a smartphone or tablet with full function USB-C.
+
+  This binding only describes the HDMI to DP display bridge.
+
+properties:
+  compatible:
+    const: analogix,anx7688
+
+  reg:
+    maxItems: 1
+    description: I2C address of the device
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object
+        description: |
+          Video port for HDMI input
+
+      port@1:
+        type: object
+        description: |
+          Video port for eDP output
+
+    required:
+      - port@0
+
+required:
+  - compatible
+  - reg
+  - ports
+
+examples:
+  - |
+    anx7688: anx7688@2c {
+      compatible = "analogix,anx7688";
+      reg = <0x2c>;
+
+      port {
+        anx7688_in: endpoint {
+          remote-endpoint = <&hdmi0_out>;
+        };
+      };
+    };
Signature · 1 line
2.24.0.525.g8f36a354ae-goog
maintainer
#2 lore raw

Re: [PATCH RESEND 1/4] dt-bindings: drm/bridge: analogix-anx7688: Add ANX7688 transmitter binding

Quoted 52 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add support for analogix,anx7688

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/anx7688.yaml      | 60 +++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/anx7688.yaml
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +32 -0
new file mode 100644
index 000000000000..cf79f7cf8fdf
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/anx7688.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Analogix ANX7688 SlimPort (Single-Chip Transmitter for DP over USB-C)
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  The ANX7688 is a single-chip mobile transmitter to support 4K 60 frames per
+  second (4096x2160p60) or FHD 120 frames per second (1920x1080p120) video
+  resolution from a smartphone or tablet with full function USB-C.
+
+  This binding only describes the HDMI to DP display bridge.
+
+properties:
+  compatible:
+    const: analogix,anx7688
+
+  reg:
+    maxItems: 1
+    description: I2C address of the device
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 42

Quoted 10 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +10 -0
+        description: |
+          Video port for HDMI input
+
+      port@1:
+        type: object
+        description: |
+          Video port for eDP output
+
+    required:
+      - port@0

Sometimes you have no output?

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 47

Quoted 5 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +5 -0
+
+required:
+  - compatible
+  - reg
+  - ports

The example will have errors because it is missing 'ports'. Run 'make dt_binding_check'.

Add:

additionalProperties: false

Quoted 16 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +13 -0
+
+examples:
+  - |
+    anx7688: anx7688@2c {
+      compatible = "analogix,anx7688";
+      reg = <0x2c>;
+
+      port {
+        anx7688_in: endpoint {
+          remote-endpoint = <&hdmi0_out>;
+        };
+      };
+    };
Signature · 1 line
2.24.0.525.g8f36a354ae-goog

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#3 lore raw

Re: [PATCH RESEND 1/4] dt-bindings: drm/bridge: analogix-anx7688: Add ANX7688 transmitter binding

Quoted 58 lines · Rob Herring
Quoted 56 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add support for analogix,anx7688

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/anx7688.yaml      | 60 +++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/anx7688.yaml
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +36 -0
new file mode 100644
index 000000000000..cf79f7cf8fdf
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/anx7688.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Analogix ANX7688 SlimPort (Single-Chip Transmitter for DP over USB-C)
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  The ANX7688 is a single-chip mobile transmitter to support 4K 60 frames per
+  second (4096x2160p60) or FHD 120 frames per second (1920x1080p120) video
+  resolution from a smartphone or tablet with full function USB-C.
+
+  This binding only describes the HDMI to DP display bridge.
+
+properties:
+  compatible:
+    const: analogix,anx7688
+
+  reg:
+    maxItems: 1
+    description: I2C address of the device
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object
+        description: |
+          Video port for HDMI input
+
+      port@1:
Quoted 8 lines

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 42

Quoted 6 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +6 -0
+        type: object
+        description: |
+          Video port for eDP output
+
+    required:
+      - port@0

Sometimes you have no output?

Yes, only input is required.

Quoted 10 lines

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 47

Quoted 6 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +5 -0
+
+required:
+  - compatible
+  - reg
+  - ports

The example will have errors because it is missing 'ports'. Run 'make dt_binding_check'.

Quoted 4 lines

Add:

additionalProperties: false

Ack, will fix this. Thanks

Quoted 16 lines
Quoted 16 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +13 -1
+
+examples:
+  - |
+    anx7688: anx7688@2c {
+      compatible = "analogix,anx7688";
+      reg = <0x2c>;
+
+      port {
+        anx7688_in: endpoint {
+          remote-endpoint = <&hdmi0_out>;
+        };
+      };
+    };
--

2.24.0.525.g8f36a354ae-goog

#4 lore raw

Re: [PATCH RESEND 1/4] dt-bindings: drm/bridge: analogix-anx7688: Add ANX7688 transmitter binding

Hi Hsin-Yi,

Quoted 67 lines · Hsin-Yi Wang
Quoted 57 lines · Rob Herring
Quoted 56 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add support for analogix,anx7688

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/anx7688.yaml      | 60 +++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/anx7688.yaml
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +36 -0
new file mode 100644
index 000000000000..cf79f7cf8fdf
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/anx7688.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Analogix ANX7688 SlimPort (Single-Chip Transmitter for DP over USB-C)
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  The ANX7688 is a single-chip mobile transmitter to support 4K 60 frames per
+  second (4096x2160p60) or FHD 120 frames per second (1920x1080p120) video
+  resolution from a smartphone or tablet with full function USB-C.
+
+  This binding only describes the HDMI to DP display bridge.
+
+properties:
+  compatible:
+    const: analogix,anx7688
+
+  reg:
+    maxItems: 1
+    description: I2C address of the device
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object
+        description: |
+          Video port for HDMI input
+
+      port@1:
Quoted 8 lines

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 42

Quoted 6 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +6 -0
+        type: object
+        description: |
+          Video port for eDP output
+
+    required:
+      - port@0

Sometimes you have no output?

Quoted 1 line

Yes, only input is required.

But what happens in that case ? What's the use of a bridge with a non-connected output ? :-)

Quoted 29 lines
Quoted 9 lines

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 47

Quoted 5 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +5 -0
+
+required:
+  - compatible
+  - reg
+  - ports

The example will have errors because it is missing 'ports'. Run 'make dt_binding_check'.

Quoted 4 lines

Add:

additionalProperties: false

Ack, will fix this. Thanks

Quoted 13 lines
Quoted 13 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +13 -0
+
+examples:
+  - |
+    anx7688: anx7688@2c {
+      compatible = "analogix,anx7688";
+      reg = <0x2c>;
+
+      port {
+        anx7688_in: endpoint {
+          remote-endpoint = <&hdmi0_out>;
+        };
+      };
+    };
Signature · 3 lines
Regards,

Laurent Pinchart
#5 lore raw

Re: [PATCH RESEND 1/4] dt-bindings: drm/bridge: analogix-anx7688: Add ANX7688 transmitter binding

Quoted 71 lines · Laurent Pinchart

Hi Hsin-Yi,

Quoted 67 lines · Hsin-Yi Wang
Quoted 65 lines · Rob Herring
Quoted 52 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add support for analogix,anx7688

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/anx7688.yaml      | 60 +++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/anx7688.yaml
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +32 -0
new file mode 100644
index 000000000000..cf79f7cf8fdf
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/anx7688.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Analogix ANX7688 SlimPort (Single-Chip Transmitter for DP over USB-C)
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  The ANX7688 is a single-chip mobile transmitter to support 4K 60 frames per
+  second (4096x2160p60) or FHD 120 frames per second (1920x1080p120) video
+  resolution from a smartphone or tablet with full function USB-C.
+
+  This binding only describes the HDMI to DP display bridge.
+
+properties:
+  compatible:
+    const: analogix,anx7688
+
+  reg:
+    maxItems: 1
+    description: I2C address of the device
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 42

Quoted 10 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +10 -0
+        description: |
+          Video port for HDMI input
+
+      port@1:
+        type: object
+        description: |
+          Video port for eDP output
+
+    required:
+      - port@0

Sometimes you have no output?

Quoted 5 lines
Quoted 1 line

Yes, only input is required.

But what happens in that case ? What's the use of a bridge with a non-connected output ? :-)

There's output connected, but doesn't need a driver. For example, in
our use case it's connected to a usb-c connector. We don't need to
state it in dts.
I also checked https://elixir.bootlin.com/linux/v5.5-rc2/source/Documentation/devicetree/bindings/display/bridge/anx6345.yaml
and thought that output could be optional. Also
https://elixir.bootlin.com/linux/v5.5-rc2/source/arch/arm64/boot/dts/allwinner/sun50i-a64-teres-i.dts#L116
Or maybe we can add output in example as anx6345?

Quoted 34 lines
Quoted 16 lines
Quoted 9 lines

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 47

Quoted 5 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +5 -0
+
+required:
+  - compatible
+  - reg
+  - ports

The example will have errors because it is missing 'ports'. Run 'make dt_binding_check'.

Quoted 4 lines

Add:

additionalProperties: false

Ack, will fix this. Thanks

Quoted 13 lines
Quoted 5 lines
Quoted 5 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +5 -0
+
+examples:
+  - |
+    anx7688: anx7688@2c {
+      compatible = "analogix,anx7688";
Quoted 8 lines

Documentation/devicetree/bindings/display/bridge/anx7688.yaml line 60

Quoted 8 lines
Documentation/devicetree/bindings/display/bridge/anx7688.yaml new file +8 -0
+      reg = <0x2c>;
+
+      port {
+        anx7688_in: endpoint {
+          remote-endpoint = <&hdmi0_out>;
+        };
+      };
+    };

--
Regards,

Laurent Pinchart

#6 lore raw

PATCH RESEND 2/4 drm: bridge: anx7688: Add anx7688 bridge driver support.

In reply to Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
 	depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+	tristate "Analogix ANX7688 bridge"
+	select DRM_KMS_HELPER
+	select REGMAP_I2C
+	---help---
+	  ANX7688 is a transmitter to support DisplayPort over USB-C for
+	  smartphone and tablets.
+	  This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
 	tristate "Analogix ANX78XX bridge"
 	select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +202 -0

202 lines not on this page · raw message

Signature · 1 line
2.24.0.525.g8f36a354ae-goog
Tested-by on v1 2/4
#7 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Hi Hsin-Yi,

Quoted 5 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

Quoted 8 lines

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

---

Although I am not a drm expert I did an initial review of this patch before sending and looks good to me now. Also I just tested with current mainline on my ELM device and I am happy to have display now, so thanks for sending this upstream:

Quoted 245 lines
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
 	depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+	tristate "Analogix ANX7688 bridge"
+	select DRM_KMS_HELPER
+	select REGMAP_I2C
+	---help---
+	  ANX7688 is a transmitter to support DisplayPort over USB-C for
+	  smartphone and tablets.
+	  This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
 	tristate "Analogix ANX78XX bridge"
 	select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +202 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86
+
+#define VENDOR_ID 0x1f29
+#define DEVICE_ID 0x7688
+
+/* First supported firmware version (0.85) */
+#define MINIMUM_FW_VERSION 0x0085
+
+struct anx7688 {
+	struct drm_bridge bridge;
+	struct i2c_client *client;
+	struct regmap *regmap;
+
+	bool filter;
+};
+
+static inline struct anx7688 *bridge_to_anx7688(struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct anx7688, bridge);
+}
+
+static bool anx7688_bridge_mode_fixup(struct drm_bridge *bridge,
+				      const struct drm_display_mode *mode,
+				      struct drm_display_mode *adjusted_mode)
+{
+	struct anx7688 *anx7688 = bridge_to_anx7688(bridge);
+	u8 regs[2];
+	u8 dpbw, lanecount;
+	int totalbw, requiredbw;
+	int ret;
+
+	if (!anx7688->filter)
+		return true;
+
+	/* Read both regs 0x85 (bandwidth) and 0x86 (lane count). */
+	ret = regmap_bulk_read(anx7688->regmap, DP_BANDWIDTH_REG, regs, 2);
+	if (ret < 0) {
+		dev_err(&anx7688->client->dev,
+			"Failed to read bandwidth/lane count\n");
+		return false;
+	}
+	dpbw = regs[0];
+	lanecount = regs[1];
+
+	/* Maximum 0x19 bandwidth (6.75 Gbps Turbo mode), 2 lanes */
+	if (dpbw > 0x19 || lanecount > 2) {
+		dev_err(&anx7688->client->dev,
+			"Invalid bandwidth/lane count (%02x/%d)\n",
+			dpbw, lanecount);
+		return false;
+	}
+
+	/* Compute available bandwidth (kHz) */
+	totalbw = dpbw * lanecount * 270000 * 8 / 10;
+
+	/* Required bandwidth (8 bpc, kHz) */
+	requiredbw = mode->clock * 8 * 3;
+
+	dev_dbg(&anx7688->client->dev,
+		"DP bandwidth: %d kHz (%02x/%d); mode requires %d Khz\n",
+		totalbw, dpbw, lanecount, requiredbw);
+
+	if (totalbw == 0) {
+		dev_warn(&anx7688->client->dev,
+			 "Bandwidth/lane count are 0, not rejecting modes\n");
+		return true;
+	}
+
+	return totalbw >= requiredbw;
+}
+
+static const struct drm_bridge_funcs anx7688_bridge_funcs = {
+	.mode_fixup	= anx7688_bridge_mode_fixup,
+};
+
+static const struct regmap_config anx7688_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+};
+
+static int anx7688_i2c_probe(struct i2c_client *client,
+			     const struct i2c_device_id *id)
+{
+	struct anx7688 *anx7688;
+	struct device *dev = &client->dev;
+	int ret;
+	u8 buffer[4];
+	u16 vendor, device, fwversion;
+
+	anx7688 = devm_kzalloc(dev, sizeof(*anx7688), GFP_KERNEL);
+	if (!anx7688)
+		return -ENOMEM;
+
+#if IS_ENABLED(CONFIG_OF)
+	anx7688->bridge.of_node = client->dev.of_node;
+#endif
+
+	anx7688->client = client;
+	i2c_set_clientdata(client, anx7688);
+
+	anx7688->regmap =
+		devm_regmap_init_i2c(client, &anx7688_regmap_config);
+
+	/* Read both vendor and device id (4 bytes). */
+	ret = regmap_bulk_read(anx7688->regmap, VENDOR_ID_REG, buffer, 4);
+	if (ret) {
+		dev_err(dev, "Failed to read chip vendor/device id\n");
+		return ret;
+	}
+
+	vendor = (u16)buffer[1] << 8 | buffer[0];
+	device = (u16)buffer[3] << 8 | buffer[2];
+	if (vendor != VENDOR_ID || device != DEVICE_ID) {
+		dev_err(dev, "Invalid vendor/device id %04x/%04x\n",
+			vendor, device);
+		return -ENODEV;
+	}
+
+	ret = regmap_bulk_read(anx7688->regmap, FW_VERSION_REG, buffer, 2);
+	if (ret) {
+		dev_err(&client->dev, "Failed to read firmware version\n");
+		return ret;
+	}
+
+	fwversion = (u16)buffer[0] << 8 | buffer[1];
+	dev_info(dev, "ANX7688 firwmare version %02x.%02x\n",
+		 buffer[0], buffer[1]);
+
+	/* FW version >= 0.85 supports bandwidth/lane count registers */
+	if (fwversion >= MINIMUM_FW_VERSION) {
+		anx7688->filter = true;
+	} else {
+		/* Warn, but not fail, for backwards compatibility. */
+		dev_warn(dev,
+			 "Old ANX7688 FW version (%02x.%02x), not filtering\n",
+			 buffer[0], buffer[1]);
+	}
+
+	anx7688->bridge.funcs = &anx7688_bridge_funcs;
+	drm_bridge_add(&anx7688->bridge);
+
+	return 0;
+}
+
+static int anx7688_i2c_remove(struct i2c_client *client)
+{
+	struct anx7688 *anx7688 = i2c_get_clientdata(client);
+
+	drm_bridge_remove(&anx7688->bridge);
+
+	return 0;
+}
+
+static const struct i2c_device_id anx7688_id[] = {
+	{ "anx7688", 0 },
+	{ /* sentinel */ }
+};
+
+MODULE_DEVICE_TABLE(i2c, anx7688_id);
+
+#if IS_ENABLED(CONFIG_OF)
+static const struct of_device_id anx7688_match_table[] = {
+	{ .compatible = "analogix,anx7688", },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, anx7688_match_table);
+#endif
+
+static struct i2c_driver anx7688_driver = {
+	.driver = {
+		   .name = "anx7688",
+		   .of_match_table = of_match_ptr(anx7688_match_table),
+		  },
+	.probe = anx7688_i2c_probe,
+	.remove = anx7688_i2c_remove,
+	.id_table = anx7688_id,
+};
+
+module_i2c_driver(anx7688_driver);
+
+MODULE_DESCRIPTION("ANX7688 SlimPort Transmitter driver");
+MODULE_AUTHOR("Nicolas Boichat <[email protected]>");
+MODULE_LICENSE("GPL v2");
#8 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

In reply to Hsin-Yi Wang

Hi Hsin-Yi and Nicolas,

Thank you for the patch.

Quoted 9 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Quoted 55 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
 	depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+	tristate "Analogix ANX7688 bridge"
+	select DRM_KMS_HELPER
+	select REGMAP_I2C
+	---help---
+	  ANX7688 is a transmitter to support DisplayPort over USB-C for
+	  smartphone and tablets.
+	  This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
 	tristate "Analogix ANX78XX bridge"
 	select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +10 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 10 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +10 -0
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

Quoted 182 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +182 -0
+
+#define VENDOR_ID 0x1f29
+#define DEVICE_ID 0x7688
+
+/* First supported firmware version (0.85) */
+#define MINIMUM_FW_VERSION 0x0085
+
+struct anx7688 {
+	struct drm_bridge bridge;
+	struct i2c_client *client;
+	struct regmap *regmap;
+
+	bool filter;
+};
+
+static inline struct anx7688 *bridge_to_anx7688(struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct anx7688, bridge);
+}
+
+static bool anx7688_bridge_mode_fixup(struct drm_bridge *bridge,
+				      const struct drm_display_mode *mode,
+				      struct drm_display_mode *adjusted_mode)
+{
+	struct anx7688 *anx7688 = bridge_to_anx7688(bridge);
+	u8 regs[2];
+	u8 dpbw, lanecount;
+	int totalbw, requiredbw;
+	int ret;
+
+	if (!anx7688->filter)
+		return true;
+
+	/* Read both regs 0x85 (bandwidth) and 0x86 (lane count). */
+	ret = regmap_bulk_read(anx7688->regmap, DP_BANDWIDTH_REG, regs, 2);
+	if (ret < 0) {
+		dev_err(&anx7688->client->dev,
+			"Failed to read bandwidth/lane count\n");
+		return false;
+	}
+	dpbw = regs[0];
+	lanecount = regs[1];
+
+	/* Maximum 0x19 bandwidth (6.75 Gbps Turbo mode), 2 lanes */
+	if (dpbw > 0x19 || lanecount > 2) {
+		dev_err(&anx7688->client->dev,
+			"Invalid bandwidth/lane count (%02x/%d)\n",
+			dpbw, lanecount);
+		return false;
+	}
+
+	/* Compute available bandwidth (kHz) */
+	totalbw = dpbw * lanecount * 270000 * 8 / 10;
+
+	/* Required bandwidth (8 bpc, kHz) */
+	requiredbw = mode->clock * 8 * 3;
+
+	dev_dbg(&anx7688->client->dev,
+		"DP bandwidth: %d kHz (%02x/%d); mode requires %d Khz\n",
+		totalbw, dpbw, lanecount, requiredbw);
+
+	if (totalbw == 0) {
+		dev_warn(&anx7688->client->dev,
+			 "Bandwidth/lane count are 0, not rejecting modes\n");
+		return true;
+	}
+
+	return totalbw >= requiredbw;
+}
+
+static const struct drm_bridge_funcs anx7688_bridge_funcs = {
+	.mode_fixup	= anx7688_bridge_mode_fixup,
+};
+
+static const struct regmap_config anx7688_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+};
+
+static int anx7688_i2c_probe(struct i2c_client *client,
+			     const struct i2c_device_id *id)
+{
+	struct anx7688 *anx7688;
+	struct device *dev = &client->dev;
+	int ret;
+	u8 buffer[4];
+	u16 vendor, device, fwversion;
+
+	anx7688 = devm_kzalloc(dev, sizeof(*anx7688), GFP_KERNEL);
+	if (!anx7688)
+		return -ENOMEM;
+
+#if IS_ENABLED(CONFIG_OF)
+	anx7688->bridge.of_node = client->dev.of_node;
+#endif
+
+	anx7688->client = client;
+	i2c_set_clientdata(client, anx7688);
+
+	anx7688->regmap =
+		devm_regmap_init_i2c(client, &anx7688_regmap_config);
+
+	/* Read both vendor and device id (4 bytes). */
+	ret = regmap_bulk_read(anx7688->regmap, VENDOR_ID_REG, buffer, 4);
+	if (ret) {
+		dev_err(dev, "Failed to read chip vendor/device id\n");
+		return ret;
+	}
+
+	vendor = (u16)buffer[1] << 8 | buffer[0];
+	device = (u16)buffer[3] << 8 | buffer[2];
+	if (vendor != VENDOR_ID || device != DEVICE_ID) {
+		dev_err(dev, "Invalid vendor/device id %04x/%04x\n",
+			vendor, device);
+		return -ENODEV;
+	}
+
+	ret = regmap_bulk_read(anx7688->regmap, FW_VERSION_REG, buffer, 2);
+	if (ret) {
+		dev_err(&client->dev, "Failed to read firmware version\n");
+		return ret;
+	}
+
+	fwversion = (u16)buffer[0] << 8 | buffer[1];
+	dev_info(dev, "ANX7688 firwmare version %02x.%02x\n",
+		 buffer[0], buffer[1]);
+
+	/* FW version >= 0.85 supports bandwidth/lane count registers */
+	if (fwversion >= MINIMUM_FW_VERSION) {
+		anx7688->filter = true;
+	} else {
+		/* Warn, but not fail, for backwards compatibility. */
+		dev_warn(dev,
+			 "Old ANX7688 FW version (%02x.%02x), not filtering\n",
+			 buffer[0], buffer[1]);
+	}
+
+	anx7688->bridge.funcs = &anx7688_bridge_funcs;
+	drm_bridge_add(&anx7688->bridge);
+
+	return 0;
+}
+
+static int anx7688_i2c_remove(struct i2c_client *client)
+{
+	struct anx7688 *anx7688 = i2c_get_clientdata(client);
+
+	drm_bridge_remove(&anx7688->bridge);
+
+	return 0;
+}
+
+static const struct i2c_device_id anx7688_id[] = {
+	{ "anx7688", 0 },
+	{ /* sentinel */ }
+};
+
+MODULE_DEVICE_TABLE(i2c, anx7688_id);
+
+#if IS_ENABLED(CONFIG_OF)
+static const struct of_device_id anx7688_match_table[] = {
+	{ .compatible = "analogix,anx7688", },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, anx7688_match_table);
+#endif
+
+static struct i2c_driver anx7688_driver = {
+	.driver = {
+		   .name = "anx7688",
+		   .of_match_table = of_match_ptr(anx7688_match_table),
+		  },
+	.probe = anx7688_i2c_probe,
+	.remove = anx7688_i2c_remove,
+	.id_table = anx7688_id,
+};
+
+module_i2c_driver(anx7688_driver);
+
+MODULE_DESCRIPTION("ANX7688 SlimPort Transmitter driver");
+MODULE_AUTHOR("Nicolas Boichat <[email protected]>");
+MODULE_LICENSE("GPL v2");
Signature · 7 lines
Regards,

Laurent Pinchart
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
#9 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Quoted 11 lines · Laurent Pinchart

Hi Hsin-Yi and Nicolas,

Thank you for the patch.

Quoted 5 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

Quoted 8 lines
Quoted 4 lines

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

Quoted 60 lines
Quoted 60 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over USB-C for
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +15 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
Quoted 9 lines

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 5 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +5 -0
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

By firmware developed by ANX provided to Google.

Quoted 187 lines
Quoted 172 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +172 -0
+
+#define VENDOR_ID 0x1f29
+#define DEVICE_ID 0x7688
+
+/* First supported firmware version (0.85) */
+#define MINIMUM_FW_VERSION 0x0085
+
+struct anx7688 {
+     struct drm_bridge bridge;
+     struct i2c_client *client;
+     struct regmap *regmap;
+
+     bool filter;
+};
+
+static inline struct anx7688 *bridge_to_anx7688(struct drm_bridge *bridge)
+{
+     return container_of(bridge, struct anx7688, bridge);
+}
+
+static bool anx7688_bridge_mode_fixup(struct drm_bridge *bridge,
+                                   const struct drm_display_mode *mode,
+                                   struct drm_display_mode *adjusted_mode)
+{
+     struct anx7688 *anx7688 = bridge_to_anx7688(bridge);
+     u8 regs[2];
+     u8 dpbw, lanecount;
+     int totalbw, requiredbw;
+     int ret;
+
+     if (!anx7688->filter)
+             return true;
+
+     /* Read both regs 0x85 (bandwidth) and 0x86 (lane count). */
+     ret = regmap_bulk_read(anx7688->regmap, DP_BANDWIDTH_REG, regs, 2);
+     if (ret < 0) {
+             dev_err(&anx7688->client->dev,
+                     "Failed to read bandwidth/lane count\n");
+             return false;
+     }
+     dpbw = regs[0];
+     lanecount = regs[1];
+
+     /* Maximum 0x19 bandwidth (6.75 Gbps Turbo mode), 2 lanes */
+     if (dpbw > 0x19 || lanecount > 2) {
+             dev_err(&anx7688->client->dev,
+                     "Invalid bandwidth/lane count (%02x/%d)\n",
+                     dpbw, lanecount);
+             return false;
+     }
+
+     /* Compute available bandwidth (kHz) */
+     totalbw = dpbw * lanecount * 270000 * 8 / 10;
+
+     /* Required bandwidth (8 bpc, kHz) */
+     requiredbw = mode->clock * 8 * 3;
+
+     dev_dbg(&anx7688->client->dev,
+             "DP bandwidth: %d kHz (%02x/%d); mode requires %d Khz\n",
+             totalbw, dpbw, lanecount, requiredbw);
+
+     if (totalbw == 0) {
+             dev_warn(&anx7688->client->dev,
+                      "Bandwidth/lane count are 0, not rejecting modes\n");
+             return true;
+     }
+
+     return totalbw >= requiredbw;
+}
+
+static const struct drm_bridge_funcs anx7688_bridge_funcs = {
+     .mode_fixup     = anx7688_bridge_mode_fixup,
+};
+
+static const struct regmap_config anx7688_regmap_config = {
+     .reg_bits = 8,
+     .val_bits = 8,
+};
+
+static int anx7688_i2c_probe(struct i2c_client *client,
+                          const struct i2c_device_id *id)
+{
+     struct anx7688 *anx7688;
+     struct device *dev = &client->dev;
+     int ret;
+     u8 buffer[4];
+     u16 vendor, device, fwversion;
+
+     anx7688 = devm_kzalloc(dev, sizeof(*anx7688), GFP_KERNEL);
+     if (!anx7688)
+             return -ENOMEM;
+
+#if IS_ENABLED(CONFIG_OF)
+     anx7688->bridge.of_node = client->dev.of_node;
+#endif
+
+     anx7688->client = client;
+     i2c_set_clientdata(client, anx7688);
+
+     anx7688->regmap =
+             devm_regmap_init_i2c(client, &anx7688_regmap_config);
+
+     /* Read both vendor and device id (4 bytes). */
+     ret = regmap_bulk_read(anx7688->regmap, VENDOR_ID_REG, buffer, 4);
+     if (ret) {
+             dev_err(dev, "Failed to read chip vendor/device id\n");
+             return ret;
+     }
+
+     vendor = (u16)buffer[1] << 8 | buffer[0];
+     device = (u16)buffer[3] << 8 | buffer[2];
+     if (vendor != VENDOR_ID || device != DEVICE_ID) {
+             dev_err(dev, "Invalid vendor/device id %04x/%04x\n",
+                     vendor, device);
+             return -ENODEV;
+     }
+
+     ret = regmap_bulk_read(anx7688->regmap, FW_VERSION_REG, buffer, 2);
+     if (ret) {
+             dev_err(&client->dev, "Failed to read firmware version\n");
+             return ret;
+     }
+
+     fwversion = (u16)buffer[0] << 8 | buffer[1];
+     dev_info(dev, "ANX7688 firwmare version %02x.%02x\n",
+              buffer[0], buffer[1]);
+
+     /* FW version >= 0.85 supports bandwidth/lane count registers */
+     if (fwversion >= MINIMUM_FW_VERSION) {
+             anx7688->filter = true;
+     } else {
+             /* Warn, but not fail, for backwards compatibility. */
+             dev_warn(dev,
+                      "Old ANX7688 FW version (%02x.%02x), not filtering\n",
+                      buffer[0], buffer[1]);
+     }
+
+     anx7688->bridge.funcs = &anx7688_bridge_funcs;
+     drm_bridge_add(&anx7688->bridge);
+
+     return 0;
+}
+
+static int anx7688_i2c_remove(struct i2c_client *client)
+{
+     struct anx7688 *anx7688 = i2c_get_clientdata(client);
+
+     drm_bridge_remove(&anx7688->bridge);
+
+     return 0;
+}
+
+static const struct i2c_device_id anx7688_id[] = {
+     { "anx7688", 0 },
+     { /* sentinel */ }
+};
+
+MODULE_DEVICE_TABLE(i2c, anx7688_id);
+
+#if IS_ENABLED(CONFIG_OF)
+static const struct of_device_id anx7688_match_table[] = {
+     { .compatible = "analogix,anx7688", },
+     { /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, anx7688_match_table);
+#endif
+
+static struct i2c_driver anx7688_driver = {
+     .driver = {
+                .name = "anx7688",
+                .of_match_table = of_match_ptr(anx7688_match_table),
+               },

drivers/gpu/drm/bridge/analogix-anx7688.c line 202

Quoted 10 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +10 -0
+     .probe = anx7688_i2c_probe,
+     .remove = anx7688_i2c_remove,
+     .id_table = anx7688_id,
+};
+
+module_i2c_driver(anx7688_driver);
+
+MODULE_DESCRIPTION("ANX7688 SlimPort Transmitter driver");
+MODULE_AUTHOR("Nicolas Boichat <[email protected]>");
+MODULE_LICENSE("GPL v2");

--
Regards,

Laurent Pinchart

#10 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Quoted 19 lines · Hsin-Yi Wang
Quoted 16 lines · Laurent Pinchart

Hi Hsin-Yi and Nicolas,

Thank you for the patch.

Quoted 9 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

Quoted 6 lines
Quoted 3 lines

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

To clarify a bit more, this is because this chip is actually a TCPC+mux+HDMI=>DP converter
(https://www.analogix.com/en/products/convertersbridges/anx7688). In Chromebook architecture, TCPC+mux is controlled by the EC (including power and other control pins), and the only reason we need a driver for the HDMI=>DP converter is to get the number of lanes on the DP side and filter out resolutions. Also, the converter is on a different I2C address and it could almost be considered as a separate device.

(of course we could write a kernel driver for the TCPC+mux but we'll leave that to others if there's ever a board that is built with the TCPC part connected to the AP)

Quoted 69 lines
Quoted 60 lines
Quoted 60 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over USB-C for
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +15 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
Quoted 9 lines

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 5 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +5 -0
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

Quoted 1 line

By firmware developed by ANX provided to Google.

We asked for these registers to be added to ANX FW, and this is the FW that is used by all elm/hana Chromebooks (I have no idea about other ANX customers...). We have facilities to update the ANX FW from coreboot/depthcharge on Chromebooks, but that does not really matter:
the factory FW of all MP Chromebooks does provide these registers.

Thanks.
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#11 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Hello Nicolas and Hsin-Yi,

Quoted 29 lines · Nicolas Boichat
Quoted 12 lines · Hsin-Yi Wang
Quoted 11 lines · Laurent Pinchart
Quoted 9 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

Quoted 6 lines
Quoted 3 lines

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

To clarify a bit more, this is because this chip is actually a TCPC+mux+HDMI=>DP converter
(https://www.analogix.com/en/products/convertersbridges/anx7688). In Chromebook architecture, TCPC+mux is controlled by the EC (including power and other control pins), and the only reason we need a driver for the HDMI=>DP converter is to get the number of lanes on the DP side and filter out resolutions. Also, the converter is on a different I2C address and it could almost be considered as a separate device.

Quoted 3 lines

(of course we could write a kernel driver for the TCPC+mux but we'll leave that to others if there's ever a board that is built with the TCPC part connected to the AP)

Is the mux the one that is handled through a gpio-mux driver in this series, or a different mux ? It would really, really help if you could show a block diagram of the related hardware (including the EC), as this is quite confusing. With every e-mail exchanged there's a bit more information that change my understanding of the issue, I can't really provide guidance without a full overview.

Quoted 69 lines
Quoted 69 lines
Quoted 69 lines
Quoted 55 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over USB-C for
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +10 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 10 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +10 -0
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

Quoted 7 lines
Quoted 1 line

By firmware developed by ANX provided to Google.

We asked for these registers to be added to ANX FW, and this is the FW that is used by all elm/hana Chromebooks (I have no idea about other ANX customers...). We have facilities to update the ANX FW from coreboot/depthcharge on Chromebooks, but that does not really matter:
the factory FW of all MP Chromebooks does provide these registers.

So the driver is specific to Chromebooks, it doesn't support all ANX7688. Sweet :-(

Signature · 7 lines
Regards,

Laurent Pinchart
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
#12 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Hi Laurent,

Quoted 10 lines · Laurent Pinchart

Hello Nicolas and Hsin-Yi,

Quoted 6 lines · Nicolas Boichat
Quoted 5 lines · Hsin-Yi Wang
Quoted 4 lines · Laurent Pinchart
Quoted 3 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port

controller),

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the

bridge

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 1 line

via 2 registers on I2C.

How about power, doesn't this chip have power supplies that

potentially

Quoted 15 lines
Quoted 15 lines
Quoted 5 lines
Quoted 2 lines

need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

To clarify a bit more, this is because this chip is actually a TCPC+mux+HDMI=>DP converter
(https://www.analogix.com/en/products/convertersbridges/anx7688). In Chromebook architecture, TCPC+mux is controlled by the EC (including power and other control pins), and the only reason we need a driver for the HDMI=>DP converter is to get the number of lanes on the DP side and filter out resolutions. Also, the converter is on a different I2C address and it could almost be considered as a separate device.

Quoted 6 lines
Quoted 3 lines

(of course we could write a kernel driver for the TCPC+mux but we'll leave that to others if there's ever a board that is built with the TCPC part connected to the AP)

Is the mux the one that is handled through a gpio-mux driver in this series, or a different mux ?

It's a different mux: it's the usual USB-C mux that takes in USB 3.0 and DP (internally converted from HDMI), and decides which 2 lanes to use for each, and flip if necessary. This is all controlled by the EC (like on most other Chromebook), so this is transparent to the kernel, at least on Chromebooks.

Quoted 5 lines

It would really, really help if you could
show a block diagram of the related hardware (including the EC), as this is quite confusing. With every e-mail exchanged there's a bit more information that change my understanding of the issue, I can't really provide guidance without a full overview.

https://lkml.org/lkml/2019/12/9/548 that you drew is accurate for the display part of the problem.

You can just add a USB3 connection to the above (there's also I2C interface to the EC of course to control the TCPC/mux aspect of it, but that's on different I2C addresses).

                                     +-----------+
+---------+         +------+    /--> | HDMI      |
| MT8173  |  HDMI   |   -->| --/     | Connector |
|  HDMI   | ------> |--/   |         +-----------+
| Encoder |         |    ->| --\     +-----------+      +-----------+
+---------+         +------+    \--> | ANX7688   | ---> | USB-C     |
                                     | Bridge    |      | Connector |
                             USB3--> | + mux     |      |           |
				     +-----------+      +-----------+
                                         ^

The ANX7688 brige could _almost_ work driverless (and it does already), the _only_ thing that the driver is doing is filtering out impossible resolution based on DP (over USB-C) number of lanes and bandwidth.

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202

++++++++++++++++++++++

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

Review

Quoted 4 lines
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
(quoted diff) +0 -0

b/drivers/gpu/drm/bridge/Kconfig

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

index 34362976cd6f..1f3fc6bec842 100644

drivers/gpu/drm/bridge/Kconfig +0 -0
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines

drivers/gpu/drm/bridge/Kconfig config DRM_PANEL_BRIDGE

Quoted 6 lines
drivers/gpu/drm/bridge/Kconfig +6 -0
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over

USB-C for

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

drivers/gpu/drm/bridge/Kconfig config DRM_PANEL_BRIDGE · line 26

Quoted 2 lines
drivers/gpu/drm/bridge/Kconfig +2 -0
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the

chip.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

Review

Quoted 5 lines
drivers/gpu/drm/bridge/Kconfig +1 -0
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
(quoted diff) +0 -0

b/drivers/gpu/drm/bridge/Makefile

Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines

Review

Quoted 10 lines

index 4934fcf5a6f8..7a1e0ec032e6 100644

drivers/gpu/drm/bridge/Makefile +1 -0
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
(quoted diff) +0 -0

b/drivers/gpu/drm/bridge/analogix-anx7688.c

Quoted 19 lines
Quoted 19 lines
Quoted 19 lines
Quoted 19 lines
Quoted 19 lines

new file mode 100644
index 000000000000..baaed48d6201

drivers/gpu/drm/bridge/analogix-anx7688.c new file +14 -0
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 6 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +6 -0
+#define DEVICE_ID_REG 0x02
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the

firmware

Quoted 4 lines
Quoted 4 lines
Quoted 3 lines
Quoted 2 lines

running on the chip (and, I assume, developed by Google) ?

By firmware developed by ANX provided to Google.

Quoted 8 lines
Quoted 5 lines

We asked for these registers to be added to ANX FW, and this is the FW that is used by all elm/hana Chromebooks (I have no idea about other ANX customers...). We have facilities to update the ANX FW from coreboot/depthcharge on Chromebooks, but that does not really matter:
the factory FW of all MP Chromebooks does provide these registers.

So the driver is specific to Chromebooks, it doesn't support all ANX7688. Sweet :-(

FWIW, this is a 3+ year old part, so it appears that nobody else cares anyway?

Also, this driver is only required to implement the mode filtering, which, possibly, is only supported by the Google version of the FW (I have no idea what other customers ANX has for this part).

Quoted 4 lines

--
Regards,

Laurent Pinchart

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#13 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

(Brilliant, I managed to accidentally send the email below, and send it as HTML, sorry about that... ASCII art in gmail is hard ,-(

Take 2:)

Hi Laurent,

Quoted 34 lines · Laurent Pinchart
Quoted 33 lines

Hello Nicolas and Hsin-Yi,

Quoted 29 lines · Nicolas Boichat
Quoted 12 lines · Hsin-Yi Wang
Quoted 11 lines · Laurent Pinchart
Quoted 9 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

Quoted 6 lines
Quoted 3 lines

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

To clarify a bit more, this is because this chip is actually a TCPC+mux+HDMI=>DP converter
(https://www.analogix.com/en/products/convertersbridges/anx7688). In Chromebook architecture, TCPC+mux is controlled by the EC (including power and other control pins), and the only reason we need a driver for the HDMI=>DP converter is to get the number of lanes on the DP side and filter out resolutions. Also, the converter is on a different I2C address and it could almost be considered as a separate device.

Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

(of course we could write a kernel driver for the TCPC+mux but we'll leave that to others if there's ever a board that is built with the TCPC part connected to the AP)

Is the mux the one that is handled through a gpio-mux driver in this series, or a different mux ?

It's a different mux: it's the usual USB-C mux that takes in USB 3.0 and DP (internally converted from HDMI), and decides which 2 lanes to use for each (4 lanes in total, but DP can only take 2 with this converter), and flip if necessary. This is all controlled by the EC (like on most other Chromebooks), so this is transparent to the kernel on this hardware.

Quoted 5 lines
Quoted 5 lines

It would really, really help if you could
show a block diagram of the related hardware (including the EC), as this is quite confusing. With every e-mail exchanged there's a bit more information that change my understanding of the issue, I can't really provide guidance without a full overview.

https://lkml.org/lkml/2019/12/9/548 that you drew is accurate for the display part of the problem.

You can just add a USB3 connection to the above (there's also I2C interface to the EC of course to control the TCPC/mux aspect of it, but that's on different I2C addresses). Something like this:

                                      +-----------+
 +---------+         +------+    /--> | HDMI      |
 | MT8173  |  HDMI   |   -->| --/     | Connector |
 |  HDMI   | ------> |--/   |         +-----------+
 | Encoder |         |    ->| --\     +-----------+      +-----------+
 +---------+         +------+    \--> | ANX7688   | ---> | USB-C     |
                                      | Bridge    |      | Connector |
                              USB3--> | + mux     |      |           |
                                      +-----------+      +-----------+
                                         ^     ^
                                   (I2C) |     | (I2C)
   MT8173 (DP lane count/bw readback) -- +     + -- EC (TCPC+mux control)

Power is also fully controlled by the EC.

(the product brief has a good diagram of the internals of the ANX7688:
https://www.analogix.com/en/system/files/AA-002281-PB-6-ANX7688_Product_Brief.pdf)

The ANX7688 bridge could _almost_ work driverless (and it does already), the _only_ thing that the driver is doing is filtering out impossible resolution based on DP (over USB-C) number of lanes and bandwidth. This is required to support, for example, old monitors that may only do RBR over DP (so we can't drive the full resolution over 2 DP lanes, we'd need 4 lanes, and we need to filter out the higher resolution modes).

Quoted 71 lines
Quoted 71 lines
Quoted 71 lines
Quoted 70 lines
Quoted 60 lines
Quoted 60 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over USB-C for
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +15 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
Quoted 9 lines

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 5 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +5 -0
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

By firmware developed by ANX provided to Google.

Quoted 9 lines
Quoted 8 lines
Quoted 5 lines

We asked for these registers to be added to ANX FW, and this is the FW that is used by all elm/hana Chromebooks (I have no idea about other ANX customers...). We have facilities to update the ANX FW from coreboot/depthcharge on Chromebooks, but that does not really matter:
the factory FW of all MP Chromebooks does provide these registers.

So the driver is specific to Chromebooks, it doesn't support all ANX7688. Sweet :-(

FWIW, this is a 3+ year old part, so it appears that nobody else cares anyway?

Also, this driver is only required to implement the mode filtering, which, possibly, is only supported by the Google version of the FW (I have no idea what other customers ANX has for this part, if they care about this problem, and if so, how they solve it).
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#14 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Hi Nicolas,

Quoted 2 lines · Nicolas Boichat

(Brilliant, I managed to accidentally send the email below, and send it as HTML, sorry about that... ASCII art in gmail is hard ,-(

No worries. I have been told it's indeed painful.

Quoted 76 lines

Take 2:)

Hi Laurent,

Quoted 31 lines · Laurent Pinchart
Quoted 30 lines · Nicolas Boichat
Quoted 29 lines · Hsin-Yi Wang
Quoted 12 lines · Laurent Pinchart
Quoted 11 lines · Hsin-Yi Wang
Quoted 9 lines

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

Quoted 6 lines
Quoted 3 lines

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

To clarify a bit more, this is because this chip is actually a TCPC+mux+HDMI=>DP converter
(https://www.analogix.com/en/products/convertersbridges/anx7688). In Chromebook architecture, TCPC+mux is controlled by the EC (including power and other control pins), and the only reason we need a driver for the HDMI=>DP converter is to get the number of lanes on the DP side and filter out resolutions. Also, the converter is on a different I2C address and it could almost be considered as a separate device.

Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

(of course we could write a kernel driver for the TCPC+mux but we'll leave that to others if there's ever a board that is built with the TCPC part connected to the AP)

Is the mux the one that is handled through a gpio-mux driver in this series, or a different mux ?

It's a different mux: it's the usual USB-C mux that takes in USB 3.0 and DP (internally converted from HDMI), and decides which 2 lanes to use for each (4 lanes in total, but DP can only take 2 with this converter), and flip if necessary. This is all controlled by the EC (like on most other Chromebooks), so this is transparent to the kernel on this hardware.

Quoted 5 lines
Quoted 5 lines

It would really, really help if you could
show a block diagram of the related hardware (including the EC), as this is quite confusing. With every e-mail exchanged there's a bit more information that change my understanding of the issue, I can't really provide guidance without a full overview.

https://lkml.org/lkml/2019/12/9/548 that you drew is accurate for the display part of the problem.

You can just add a USB3 connection to the above (there's also I2C interface to the EC of course to control the TCPC/mux aspect of it, but that's on different I2C addresses). Something like this:

                                      +-----------+
 +---------+         +------+    /--> | HDMI      |
 | MT8173  |  HDMI   |   -->| --/     | Connector |
 |  HDMI   | ------> |--/   |         +-----------+
 | Encoder |         |    ->| --\     +-----------+      +-----------+
 +---------+         +------+    \--> | ANX7688   | ---> | USB-C     |
                                      | Bridge    |      | Connector |
                              USB3--> | + mux     |      |           |
                                      +-----------+      +-----------+
                                         ^     ^
                                   (I2C) |     | (I2C)
   MT8173 (DP lane count/bw readback) -- +     + -- EC (TCPC+mux control)
Quoted 1 line

Power is also fully controlled by the EC.

Could I ask you to also explain how the HDMI mux is controlled, and where the HPD-related signals for the HDMI connector and USB-C connector are routed to ?

Quoted 88 lines

(the product brief has a good diagram of the internals of the ANX7688:
https://www.analogix.com/en/system/files/AA-002281-PB-6-ANX7688_Product_Brief.pdf)

The ANX7688 bridge could _almost_ work driverless (and it does already), the _only_ thing that the driver is doing is filtering out impossible resolution based on DP (over USB-C) number of lanes and bandwidth. This is required to support, for example, old monitors that may only do RBR over DP (so we can't drive the full resolution over 2 DP lanes, we'd need 4 lanes, and we need to filter out the higher resolution modes).

Quoted 77 lines
Quoted 77 lines
Quoted 76 lines
Quoted 69 lines
Quoted 60 lines
Quoted 60 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over USB-C for
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +15 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
Quoted 9 lines

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 5 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +5 -0
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

Quoted 1 line

By firmware developed by ANX provided to Google.

We asked for these registers to be added to ANX FW, and this is the FW that is used by all elm/hana Chromebooks (I have no idea about other ANX customers...). We have facilities to update the ANX FW from coreboot/depthcharge on Chromebooks, but that does not really matter:
the factory FW of all MP Chromebooks does provide these registers.

Quoted 4 lines
Quoted 2 lines
Quoted 2 lines

So the driver is specific to Chromebooks, it doesn't support all ANX7688. Sweet :-(

FWIW, this is a 3+ year old part, so it appears that nobody else cares anyway?

That's good news :-)

Quoted 4 lines

Also, this driver is only required to implement the mode filtering, which, possibly, is only supported by the Google version of the FW (I have no idea what other customers ANX has for this part, if they care about this problem, and if so, how they solve it).

Signature · 7 lines
Regards,

Laurent Pinchart
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
#15 lore raw

Re: [PATCH RESEND 2/4] drm: bridge: anx7688: Add anx7688 bridge driver support.

Hi,

Quoted 85 lines · Laurent Pinchart

Hi Nicolas,

Quoted 2 lines · Nicolas Boichat

(Brilliant, I managed to accidentally send the email below, and send it as HTML, sorry about that... ASCII art in gmail is hard ,-(

No worries. I have been told it's indeed painful.

Quoted 76 lines

Take 2:)

Hi Laurent,

Quoted 31 lines · Laurent Pinchart
Quoted 30 lines · Nicolas Boichat
Quoted 29 lines · Hsin-Yi Wang
Quoted 12 lines · Laurent Pinchart
Quoted 11 lines · Hsin-Yi Wang
Quoted 9 lines

From: Nicolas Boichat <[email protected]>

ANX7688 is a HDMI to DP converter (as well as USB-C port controller), that has an internal microcontroller.

The only reason a Linux kernel driver is necessary is to reject resolutions that require more bandwidth than what is available on the DP side. DP bandwidth and lane count are reported by the bridge via 2 registers on I2C.

Quoted 6 lines
Quoted 3 lines

How about power, doesn't this chip have power supplies that potentially need to be controlled ?

Ideally we should add power supplies as well, but the power is supplied by ec in mt8173 oak board. And we only have this board can test this driver. If we add power supplies in driver we can't test it.

To clarify a bit more, this is because this chip is actually a TCPC+mux+HDMI=>DP converter
(https://www.analogix.com/en/products/convertersbridges/anx7688). In Chromebook architecture, TCPC+mux is controlled by the EC (including power and other control pins), and the only reason we need a driver for the HDMI=>DP converter is to get the number of lanes on the DP side and filter out resolutions. Also, the converter is on a different I2C address and it could almost be considered as a separate device.

Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

(of course we could write a kernel driver for the TCPC+mux but we'll leave that to others if there's ever a board that is built with the TCPC part connected to the AP)

Is the mux the one that is handled through a gpio-mux driver in this series, or a different mux ?

It's a different mux: it's the usual USB-C mux that takes in USB 3.0 and DP (internally converted from HDMI), and decides which 2 lanes to use for each (4 lanes in total, but DP can only take 2 with this converter), and flip if necessary. This is all controlled by the EC (like on most other Chromebooks), so this is transparent to the kernel on this hardware.

Quoted 5 lines
Quoted 5 lines

It would really, really help if you could
show a block diagram of the related hardware (including the EC), as this is quite confusing. With every e-mail exchanged there's a bit more information that change my understanding of the issue, I can't really provide guidance without a full overview.

https://lkml.org/lkml/2019/12/9/548 that you drew is accurate for the display part of the problem.

You can just add a USB3 connection to the above (there's also I2C interface to the EC of course to control the TCPC/mux aspect of it, but that's on different I2C addresses). Something like this:

                                      +-----------+
 +---------+         +------+    /--> | HDMI      |
 | MT8173  |  HDMI   |   -->| --/     | Connector |
 |  HDMI   | ------> |--/   |         +-----------+
 | Encoder |         |    ->| --\     +-----------+      +-----------+
 +---------+         +------+    \--> | ANX7688   | ---> | USB-C     |
                                      | Bridge    |      | Connector |
                              USB3--> | + mux     |      |           |
                                      +-----------+      +-----------+
                                         ^     ^
                                   (I2C) |     | (I2C)
   MT8173 (DP lane count/bw readback) -- +     + -- EC (TCPC+mux control)
Quoted 3 lines
Quoted 1 line

Power is also fully controlled by the EC.

Could I ask you to also explain how the HDMI mux is controlled,

Priority to HDMI. If the HDMI is connected (looking at its HPD signal), then the HDMI signals are routed to HDMI connector. Else HDMI is routed to ANX7688/USB-C.

Quoted 3 lines

and
where the HPD-related signals for the HDMI connector and USB-C connector are routed to ?

HPD is also muxed by the mux, between the 2 inputs.
(http://www.ti.com/lit/ds/symlink/ts3dv642.pdf, if you are curious, 9.2.3 is basically how things are wired, with one of the HPD_A/B connected to SEL2)

My memory is fading away now, but I think at some point we considered having hardware send an HPD pulse when the input changes, but decided against it (for cost/complexity reasons). So that means that if both HDMI and USB-C monitors are plugged, and you unplug HDMI, the mux would switch but you would not get an HPD pulse. That's one of the reason we need to react to edges on the mux SEL signal to ask the kernel to re-read the EDID (that's in the other driver that Hsin-Yi is trying (again) to upstream in this series).

(IIRC, that's also why HDMI HPD pulse work if both connectors are plugged, we get the edge on the SEL signal an re-read the EDID).

Thanks.

Quoted 104 lines
Quoted 88 lines

(the product brief has a good diagram of the internals of the ANX7688:
https://www.analogix.com/en/system/files/AA-002281-PB-6-ANX7688_Product_Brief.pdf)

The ANX7688 bridge could _almost_ work driverless (and it does already), the _only_ thing that the driver is doing is filtering out impossible resolution based on DP (over USB-C) number of lanes and bandwidth. This is required to support, for example, old monitors that may only do RBR over DP (so we can't drive the full resolution over 2 DP lanes, we'd need 4 lanes, and we need to filter out the higher resolution modes).

Quoted 77 lines
Quoted 77 lines
Quoted 76 lines
Quoted 69 lines
Quoted 60 lines
Quoted 60 lines
---
 drivers/gpu/drm/bridge/Kconfig            |   9 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/analogix-anx7688.c | 202 ++++++++++++++++++++++
 3 files changed, 212 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/analogix-anx7688.c
drivers/gpu/drm/bridge/Kconfig +9 -0
index 34362976cd6f..1f3fc6bec842 100644
@@ -16,6 +16,15 @@ config DRM_PANEL_BRIDGE
 menu "Display Interface Bridges"
      depends on DRM && DRM_BRIDGE
 
+config DRM_ANALOGIX_ANX7688
+     tristate "Analogix ANX7688 bridge"
+     select DRM_KMS_HELPER
+     select REGMAP_I2C
+     ---help---
+       ANX7688 is a transmitter to support DisplayPort over USB-C for
+       smartphone and tablets.
+       This driver only supports the HDMI to DP component of the chip.
+
 config DRM_ANALOGIX_ANX78XX
      tristate "Analogix ANX78XX bridge"
      select DRM_KMS_HELPER
drivers/gpu/drm/bridge/Makefile +1 -0
index 4934fcf5a6f8..7a1e0ec032e6 100644
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
drivers/gpu/drm/bridge/analogix-anx7688.c new file +15 -0
new file mode 100644
index 000000000000..baaed48d6201
@@ -0,0 +1,202 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * ANX7688 HDMI->DP bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <drm/drm_bridge.h>
+
+/* Register addresses */
+#define VENDOR_ID_REG 0x00
+#define DEVICE_ID_REG 0x02
Quoted 9 lines

drivers/gpu/drm/bridge/analogix-anx7688.c line 20

Quoted 5 lines
drivers/gpu/drm/bridge/analogix-anx7688.c new file +5 -0
+
+#define FW_VERSION_REG 0x80
+
+#define DP_BANDWIDTH_REG 0x85
+#define DP_LANE_COUNT_REG 0x86

Are these registers defined by the ANX7688 hardware, or by the firmware running on the chip (and, I assume, developed by Google) ?

Quoted 1 line

By firmware developed by ANX provided to Google.

We asked for these registers to be added to ANX FW, and this is the FW that is used by all elm/hana Chromebooks (I have no idea about other ANX customers...). We have facilities to update the ANX FW from coreboot/depthcharge on Chromebooks, but that does not really matter:
the factory FW of all MP Chromebooks does provide these registers.

Quoted 4 lines
Quoted 2 lines
Quoted 2 lines

So the driver is specific to Chromebooks, it doesn't support all ANX7688. Sweet :-(

FWIW, this is a 3+ year old part, so it appears that nobody else cares anyway?

That's good news :-)

Quoted 4 lines

Also, this driver is only required to implement the mode filtering, which, possibly, is only supported by the Google version of the FW (I have no idea what other customers ANX has for this part, if they care about this problem, and if so, how they solve it).

--
Regards,

Laurent Pinchart

#16 lore raw

PATCH RESEND 3/4 dt-bindings: drm/bridge: Add GPIO display mux binding

In reply to Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add bindings for Generic GPIO mux driver.

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/gpio-mux.yaml     | 89 +++++++++++++++++++
 1 file changed, 89 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +89 -0
new file mode 100644
index 000000000000..cef098749066
@@ -0,0 +1,89 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/gpio-mux.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Generic display mux (1 input, 2 outputs)
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  This bindings describes a simple display (e.g. HDMI) mux, that has 1
+  input, and 2 outputs. The mux status is controlled by hardware, and
+  its status is read back using a GPIO.
+
+properties:
+  compatible:
+    const: gpio-display-mux
+
+  detect-gpios:
+    maxItems: 1
+    description: GPIO that indicates the active output
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object
+        description: |
+          Video port for input.
+
+      port@1:
+        type: object
+        description: |
+          2 video ports for output.
+          The reg value in the endpoints matches the GPIO status: when
+          GPIO is asserted, endpoint with reg value <1> is selected.
+
+    required:
+      - port@0
+      - port@1
+
+required:
+  - compatible
+  - detect-gpios
+  - ports
+
+examples:
+  - |
+    hdmi_mux: hdmi_mux {
+      compatible = "gpio-display-mux";
+      status = "okay";
+      detect-gpios = <&pio 36 GPIO_ACTIVE_HIGH>;
+      pinctrl-names = "default";
+      pinctrl-0 = <&hdmi_mux_pins>;
+      ddc-i2c-bus = <&hdmiddc0>;
+
+      ports {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        port@0 { /* input */
+          reg = <0>;
+
+          hdmi_mux_in: endpoint {
+            remote-endpoint = <&hdmi0_out>;
+          };
+        };
+
+        port@1 { /* output */
+          reg = <1>;
+
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          hdmi_mux_out_anx: endpoint@0 {
+            reg = <0>;
+            remote-endpoint = <&anx7688_in>;
+          };
+
+          hdmi_mux_out_hdmi: endpoint@1 {
+            reg = <1>;
+            remote-endpoint = <&hdmi_connector_in>;
+          };
+        };
+      };
+    };
Signature · 1 line
2.24.0.525.g8f36a354ae-goog
#17 lore raw

Re: [PATCH RESEND 3/4] dt-bindings: drm/bridge: Add GPIO display mux binding

Quoted 21 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add bindings for Generic GPIO mux driver.

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/gpio-mux.yaml     | 89 +++++++++++++++++++
 1 file changed, 89 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +0 -0
new file mode 100644
index 000000000000..cef098749066

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 7

Quoted 7 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +7 -0
@@ -0,0 +1,89 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/gpio-mux.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Generic display mux (1 input, 2 outputs)

What makes it generic? Doesn't the mux chip have power supply, possibly a reset line or not, etc.? What about a mux where the GPIO controls the mux?

Generally, we avoid 'generic' bindings because h/w is rarely generic. You can have a generic driver which works on multiple devices.

Quoted 22 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +22 -0
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  This bindings describes a simple display (e.g. HDMI) mux, that has 1
+  input, and 2 outputs. The mux status is controlled by hardware, and
+  its status is read back using a GPIO.
+
+properties:
+  compatible:
+    const: gpio-display-mux
+
+  detect-gpios:
+    maxItems: 1
+    description: GPIO that indicates the active output
+
+  ports:
+    type: object
+
+    properties:
+      port@0:

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 39

Quoted 10 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +10 -0
+        type: object
+        description: |
+          Video port for input.
+
+      port@1:
+        type: object
+        description: |
+          2 video ports for output.
+          The reg value in the endpoints matches the GPIO status: when
+          GPIO is asserted, endpoint with reg value <1> is selected.

You should describe 'endpoint@0' and 'endpoint@1' here too.

Quoted 5 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +5 -0
+
+    required:
+      - port@0
+      - port@1
+

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 54

Quoted 10 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +10 -0
+required:
+  - compatible
+  - detect-gpios
+  - ports
+
+examples:
+  - |
+    hdmi_mux: hdmi_mux {
+      compatible = "gpio-display-mux";
+      status = "okay";

Don't show status in examples.

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 58

Quoted 4 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +4 -0
+      detect-gpios = <&pio 36 GPIO_ACTIVE_HIGH>;
+      pinctrl-names = "default";
+      pinctrl-0 = <&hdmi_mux_pins>;
+      ddc-i2c-bus = <&hdmiddc0>;

Not documented. Is the i2c bus muxed too? If not, then this is in the wrong place.

Quoted 32 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +31 -1
+
+      ports {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        port@0 { /* input */
+          reg = <0>;
+
+          hdmi_mux_in: endpoint {
+            remote-endpoint = <&hdmi0_out>;
+          };
+        };
+
+        port@1 { /* output */
+          reg = <1>;
+
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          hdmi_mux_out_anx: endpoint@0 {
+            reg = <0>;
+            remote-endpoint = <&anx7688_in>;
+          };
+
+          hdmi_mux_out_hdmi: endpoint@1 {
+            reg = <1>;
+            remote-endpoint = <&hdmi_connector_in>;
+          };
+        };
+      };
+    };
--
Quoted 2 lines

2.24.0.525.g8f36a354ae-goog

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#18 lore raw

Re: [PATCH RESEND 3/4] dt-bindings: drm/bridge: Add GPIO display mux binding

Quoted 35 lines · Rob Herring
Quoted 21 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add bindings for Generic GPIO mux driver.

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/gpio-mux.yaml     | 89 +++++++++++++++++++
 1 file changed, 89 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +0 -0
new file mode 100644
index 000000000000..cef098749066

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 7

Quoted 7 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +7 -0
@@ -0,0 +1,89 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/gpio-mux.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Generic display mux (1 input, 2 outputs)

What makes it generic? Doesn't the mux chip have power supply, possibly a reset line or not, etc.? What about a mux where the GPIO controls the mux?

Quoted 3 lines

Generally, we avoid 'generic' bindings because h/w is rarely generic. You can have a generic driver which works on multiple devices.

Then how about making it mt8173-oak-gpio-mux? Since this is currently only used in this board.

Quoted 26 lines
Quoted 26 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +26 -0
+
+maintainers:
+  - Nicolas Boichat <[email protected]>
+
+description: |
+  This bindings describes a simple display (e.g. HDMI) mux, that has 1
+  input, and 2 outputs. The mux status is controlled by hardware, and
+  its status is read back using a GPIO.
+
+properties:
+  compatible:
+    const: gpio-display-mux
+
+  detect-gpios:
+    maxItems: 1
+    description: GPIO that indicates the active output
+
+  ports:
+    type: object
+
+    properties:
+      port@0:
+        type: object
+        description: |
+          Video port for input.
+
Quoted 8 lines

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 39

Quoted 6 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +6 -0
+      port@1:
+        type: object
+        description: |
+          2 video ports for output.
+          The reg value in the endpoints matches the GPIO status: when
+          GPIO is asserted, endpoint with reg value <1> is selected.

You should describe 'endpoint@0' and 'endpoint@1' here too.

Will add in next version, thanks

Quoted 19 lines
Quoted 6 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +5 -0
+
+    required:
+      - port@0
+      - port@1
+

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 54

Quoted 10 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +10 -0
+required:
+  - compatible
+  - detect-gpios
+  - ports
+
+examples:
+  - |
+    hdmi_mux: hdmi_mux {
+      compatible = "gpio-display-mux";
+      status = "okay";

Don't show status in examples.

Quoted 8 lines

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 58

Quoted 4 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +4 -0
+      detect-gpios = <&pio 36 GPIO_ACTIVE_HIGH>;
+      pinctrl-names = "default";
+      pinctrl-0 = <&hdmi_mux_pins>;
+      ddc-i2c-bus = <&hdmiddc0>;

Not documented. Is the i2c bus muxed too? If not, then this is in the wrong place.

It's muxed, but this is required because of [1], so it should be removed in this example.

[1]https://elixir.bootlin.com/linux/v5.5-rc2/source/Documentation/devicetree/bindings/display/mediatek/mediatek,hdmi.txt#L24

Quoted 34 lines
Quoted 34 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +31 -1
+
+      ports {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        port@0 { /* input */
+          reg = <0>;
+
+          hdmi_mux_in: endpoint {
+            remote-endpoint = <&hdmi0_out>;
+          };
+        };
+
+        port@1 { /* output */
+          reg = <1>;
+
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          hdmi_mux_out_anx: endpoint@0 {
+            reg = <0>;
+            remote-endpoint = <&anx7688_in>;
+          };
+
+          hdmi_mux_out_hdmi: endpoint@1 {
+            reg = <1>;
+            remote-endpoint = <&hdmi_connector_in>;
+          };
+        };
+      };
+    };
--

2.24.0.525.g8f36a354ae-goog

maintainer
#19 lore raw

Re: [PATCH RESEND 3/4] dt-bindings: drm/bridge: Add GPIO display mux binding

Quoted 36 lines · Hsin-Yi Wang
Quoted 35 lines · Rob Herring
Quoted 21 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add bindings for Generic GPIO mux driver.

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/gpio-mux.yaml     | 89 +++++++++++++++++++
 1 file changed, 89 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +0 -0
new file mode 100644
index 000000000000..cef098749066

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 7

Quoted 7 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +7 -0
@@ -0,0 +1,89 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/gpio-mux.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Generic display mux (1 input, 2 outputs)

What makes it generic? Doesn't the mux chip have power supply, possibly a reset line or not, etc.? What about a mux where the GPIO controls the mux?

Quoted 5 lines
Quoted 3 lines

Generally, we avoid 'generic' bindings because h/w is rarely generic. You can have a generic driver which works on multiple devices.

Then how about making it mt8173-oak-gpio-mux? Since this is currently only used in this board.

Isn't there an underlying part# you can use? Or if you can point me to multiple chips implementing the same thing, then maybe a generic binding is fine.

Rob

#20 lore raw

Re: [PATCH RESEND 3/4] dt-bindings: drm/bridge: Add GPIO display mux binding

Quoted 41 lines · Rob Herring
Quoted 39 lines · Hsin-Yi Wang
Quoted 38 lines · Rob Herring
Quoted 21 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

Add bindings for Generic GPIO mux driver.

---
Change from RFC to v1:
- txt to yaml

---
 .../bindings/display/bridge/gpio-mux.yaml     | 89 +++++++++++++++++++
 1 file changed, 89 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +0 -0
new file mode 100644
index 000000000000..cef098749066

Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml line 7

Quoted 7 lines
Documentation/devicetree/bindings/display/bridge/gpio-mux.yaml new file +7 -0
@@ -0,0 +1,89 @@
+# SPDX-License-Identifier: GPL-2.0
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/gpio-mux.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Generic display mux (1 input, 2 outputs)

What makes it generic? Doesn't the mux chip have power supply, possibly a reset line or not, etc.? What about a mux where the GPIO controls the mux?

Generally, we avoid 'generic' bindings because h/w is rarely generic. You can have a generic driver which works on multiple devices.

Quoted 6 lines
Quoted 2 lines

Then how about making it mt8173-oak-gpio-mux? Since this is currently only used in this board.

Isn't there an underlying part# you can use? Or if you can point me to multiple chips implementing the same thing, then maybe a generic binding is fine.

There are some similar chips, for example:
https://www.paradetech.com/zh-hant/%E7%94%A2%E5%93%81%E4%BB%8B%E7%B4%B9/ps8223-3-0gbps-hdmi-12-demultiplexer/
and http://www.ti.com/lit/ds/symlink/ts3dv642.pdf
If they are used in a similar way
(https://lore.kernel.org/lkml/CANMq1KDDEzPWhByEtn-EjNcg+ofVT2MW-hOXANGooYFOYJ35VA@mail.gmail.com/),
they would need such driver. But currently we only know that mt8173 oak board have this use case.

Quoted 2 lines

Rob

#21 lore raw

PATCH RESEND 4/4 drm: bridge: Generic GPIO mux driver

In reply to Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

This driver supports single input, 2 output display mux (e.g. HDMI mux), that provide its status via a GPIO.

---
 drivers/gpu/drm/bridge/Kconfig            |  10 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/generic-gpio-mux.c | 306 ++++++++++++++++++++++
 3 files changed, 317 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/generic-gpio-mux.c
drivers/gpu/drm/bridge/Kconfig +10 -0
index 1f3fc6bec842..4734f6993858 100644
@@ -54,6 +54,16 @@ config DRM_DUMB_VGA_DAC
 	  Support for non-programmable RGB to VGA DAC bridges, such as ADI
 	  ADV7123, TI THS8134 and THS8135 or passive resistor ladder DACs.
 
+config DRM_GENERIC_GPIO_MUX
+	tristate "Generic GPIO-controlled mux"
+	depends on OF
+	select DRM_KMS_HELPER
+	---help---
+	  This bridge driver models a GPIO-controlled display mux with one
+	  input, 2 outputs (e.g. an HDMI mux). The hardware decides which output
+	  is active, reports it as a GPIO, and the driver redirects calls to the
+	  appropriate downstream bridge (if any).
+
 config DRM_LVDS_ENCODER
 	tristate "Transparent parallel to LVDS encoder support"
 	depends on OF
drivers/gpu/drm/bridge/Makefile +1 -0
index 7a1e0ec032e6..1c0c92667ac4 100644
@@ -3,6 +3,7 @@ obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
+obj-$(CONFIG_DRM_GENERIC_GPIO_MUX) += generic-gpio-mux.o
 obj-$(CONFIG_DRM_LVDS_ENCODER) += lvds-encoder.o
 obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o
 obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +306 -0

306 lines not on this page · raw message

Signature · 1 line
2.24.0.525.g8f36a354ae-goog
Tested-by on v1 4/4
#22 lore raw

Re: [PATCH RESEND 4/4] drm: bridge: Generic GPIO mux driver

Hi Hsin-Yi,

Quoted 7 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

This driver supports single input, 2 output display mux (e.g. HDMI mux), that provide its status via a GPIO.

I'll let drm maintainers comment on this but if that's the way to go you can add my:

Only one issue that needs to be solved in next version, see below.

Quoted 122 lines
---
 drivers/gpu/drm/bridge/Kconfig            |  10 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/generic-gpio-mux.c | 306 ++++++++++++++++++++++
 3 files changed, 317 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/generic-gpio-mux.c
drivers/gpu/drm/bridge/Kconfig +10 -0
index 1f3fc6bec842..4734f6993858 100644
@@ -54,6 +54,16 @@ config DRM_DUMB_VGA_DAC
 	  Support for non-programmable RGB to VGA DAC bridges, such as ADI
 	  ADV7123, TI THS8134 and THS8135 or passive resistor ladder DACs.
 
+config DRM_GENERIC_GPIO_MUX
+	tristate "Generic GPIO-controlled mux"
+	depends on OF
+	select DRM_KMS_HELPER
+	---help---
+	  This bridge driver models a GPIO-controlled display mux with one
+	  input, 2 outputs (e.g. an HDMI mux). The hardware decides which output
+	  is active, reports it as a GPIO, and the driver redirects calls to the
+	  appropriate downstream bridge (if any).
+
 config DRM_LVDS_ENCODER
 	tristate "Transparent parallel to LVDS encoder support"
 	depends on OF
drivers/gpu/drm/bridge/Makefile +1 -0
index 7a1e0ec032e6..1c0c92667ac4 100644
@@ -3,6 +3,7 @@ obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
+obj-$(CONFIG_DRM_GENERIC_GPIO_MUX) += generic-gpio-mux.o
 obj-$(CONFIG_DRM_LVDS_ENCODER) += lvds-encoder.o
 obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o
 obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +76 -0
new file mode 100644
index 000000000000..ba08321dcc17
@@ -0,0 +1,306 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Generic gpio mux bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/of_graph.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
+
+struct gpio_display_mux {
+	struct device *dev;
+
+	struct gpio_desc *gpiod_detect;
+	int detect_irq;
+
+	struct drm_bridge bridge;
+
+	struct drm_bridge *next[2];
+};
+
+static inline struct gpio_display_mux *bridge_to_gpio_display_mux(
+		struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct gpio_display_mux, bridge);
+}
+
+static irqreturn_t gpio_display_mux_det_threaded_handler(int unused, void *data)
+{
+	struct gpio_display_mux *gpio_display_mux = data;
+	int active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+
+	dev_dbg(gpio_display_mux->dev, "Interrupt %d!\n", active);
+
+	if (gpio_display_mux->bridge.dev)
+		drm_kms_helper_hotplug_event(gpio_display_mux->bridge.dev);
+
+	return IRQ_HANDLED;
+}
+
+static int gpio_display_mux_attach(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+			bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next)
+			next->encoder = bridge->encoder;
+	}
+
+	return 0;
+}
+
+static bool gpio_display_mux_mode_fixup(struct drm_bridge *bridge,
+				const struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	int active;
+	struct drm_bridge *next;
+
+	active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+	next = gpio_display_mux->next[active];

drivers/gpu/drm/bridge/generic-gpio-mux.c line 86

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+
+	if (next && next->funcs->mode_fixup)
+		return next->funcs->mode_fixup(next, mode, adjusted_mode);
+	else
+		return true;
+}
+
+static void gpio_display_mux_mode_set(struct drm_bridge *bridge,
+				struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)

Those two need to be const now.

Quoted 221 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +220 -0
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	int active;
+	struct drm_bridge *next;
+
+	active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+	next = gpio_display_mux->next[active];
+
+	if (next && next->funcs->mode_set)
+		next->funcs->mode_set(next, mode, adjusted_mode);
+}
+
+/**
+ * Since this driver _reacts_ to mux changes, we need to make sure all
+ * downstream bridges are pre-enabled.
+ */
+static void gpio_display_mux_pre_enable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->pre_enable)
+			next->funcs->pre_enable(next);
+	}
+}
+
+static void gpio_display_mux_post_disable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->post_disable)
+			next->funcs->post_disable(next);
+	}
+}
+
+/**
+ * In an ideal mux driver, only the currently selected bridge should be enabled.
+ * For the sake of simplicity, we just just enable/disable all downstream
+ * bridges at the same time.
+ */
+static void gpio_display_mux_enable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->enable)
+			next->funcs->enable(next);
+	}
+}
+
+static void gpio_display_mux_disable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->disable)
+			next->funcs->disable(next);
+	}
+}
+
+static const struct drm_bridge_funcs gpio_display_mux_bridge_funcs = {
+	.attach = gpio_display_mux_attach,
+	.mode_fixup = gpio_display_mux_mode_fixup,
+	.disable = gpio_display_mux_disable,
+	.post_disable = gpio_display_mux_post_disable,
+	.mode_set = gpio_display_mux_mode_set,
+	.pre_enable = gpio_display_mux_pre_enable,
+	.enable = gpio_display_mux_enable,
+};
+
+static int gpio_display_mux_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct gpio_display_mux *gpio_display_mux;
+	struct device_node *port, *ep, *remote;
+	int ret;
+	u32 reg;
+
+	gpio_display_mux = devm_kzalloc(dev, sizeof(*gpio_display_mux),
+					GFP_KERNEL);
+	if (!gpio_display_mux)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, gpio_display_mux);
+	gpio_display_mux->dev = &pdev->dev;
+
+	gpio_display_mux->bridge.of_node = dev->of_node;
+
+	gpio_display_mux->gpiod_detect =
+		devm_gpiod_get(dev, "detect", GPIOD_IN);
+	if (IS_ERR(gpio_display_mux->gpiod_detect))
+		return PTR_ERR(gpio_display_mux->gpiod_detect);
+
+	gpio_display_mux->detect_irq =
+		gpiod_to_irq(gpio_display_mux->gpiod_detect);
+	if (gpio_display_mux->detect_irq < 0) {
+		dev_err(dev, "Failed to get output irq %d\n",
+			gpio_display_mux->detect_irq);
+		return -ENODEV;
+	}
+
+	port = of_graph_get_port_by_id(dev->of_node, 1);
+	if (!port) {
+		dev_err(dev, "Missing output port node\n");
+		return -EINVAL;
+	}
+
+	for_each_child_of_node(port, ep) {
+		if (!ep->name || (of_node_cmp(ep->name, "endpoint") != 0)) {
+			of_node_put(ep);
+			continue;
+		}
+
+		if (of_property_read_u32(ep, "reg", &reg) < 0 ||
+				reg >= ARRAY_SIZE(gpio_display_mux->next)) {
+			dev_err(dev,
+			    "Missing/invalid reg property for endpoint %s\n",
+				ep->full_name);
+			of_node_put(ep);
+			of_node_put(port);
+			return -EINVAL;
+		}
+
+		remote = of_graph_get_remote_port_parent(ep);
+		if (!remote) {
+			dev_err(dev,
+			    "Missing connector/bridge node for endpoint %s\n",
+				ep->full_name);
+			of_node_put(ep);
+			of_node_put(port);
+			return -EINVAL;
+		}
+		of_node_put(ep);
+
+		if (of_device_is_compatible(remote, "hdmi-connector")) {
+			of_node_put(remote);
+			continue;
+		}
+
+		gpio_display_mux->next[reg] = of_drm_find_bridge(remote);
+		if (!gpio_display_mux->next[reg]) {
+			dev_err(dev, "Waiting for external bridge %s\n",
+				remote->name);
+			of_node_put(remote);
+			of_node_put(port);
+			return -EPROBE_DEFER;
+		}
+
+		of_node_put(remote);
+	}
+	of_node_put(port);
+
+	gpio_display_mux->bridge.funcs = &gpio_display_mux_bridge_funcs;
+	drm_bridge_add(&gpio_display_mux->bridge);
+
+	ret = devm_request_threaded_irq(dev, gpio_display_mux->detect_irq,
+				NULL,
+				gpio_display_mux_det_threaded_handler,
+				IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING |
+					IRQF_ONESHOT,
+				"gpio-display-mux-det", gpio_display_mux);
+	if (ret) {
+		dev_err(dev, "Failed to request MUX_DET threaded irq\n");
+		goto err_bridge_remove;
+	}
+
+	return 0;
+
+err_bridge_remove:
+	drm_bridge_remove(&gpio_display_mux->bridge);
+
+	return ret;
+}
+
+static int gpio_display_mux_remove(struct platform_device *pdev)
+{
+	struct gpio_display_mux *gpio_display_mux = platform_get_drvdata(pdev);
+
+	drm_bridge_remove(&gpio_display_mux->bridge);
+
+	return 0;
+}
+
+static const struct of_device_id gpio_display_mux_match[] = {
+	{ .compatible = "gpio-display-mux", },
+	{},
+};
+
+struct platform_driver gpio_display_mux_driver = {
+	.probe = gpio_display_mux_probe,
+	.remove = gpio_display_mux_remove,
+	.driver = {
+		.name = "gpio-display-mux",
+		.of_match_table = gpio_display_mux_match,
+	},
+};
+
+module_platform_driver(gpio_display_mux_driver);
+
+MODULE_DESCRIPTION("GPIO-controlled display mux");
+MODULE_AUTHOR("Nicolas Boichat <[email protected]>");
+MODULE_LICENSE("GPL v2");
#23 lore raw

Re: [PATCH RESEND 4/4] drm: bridge: Generic GPIO mux driver

In reply to Hsin-Yi Wang

Hi Hsin-Yi and Nicolas,

Thank you for the patch.

Quoted 21 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

This driver supports single input, 2 output display mux (e.g. HDMI mux), that provide its status via a GPIO.

---
 drivers/gpu/drm/bridge/Kconfig            |  10 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/generic-gpio-mux.c | 306 ++++++++++++++++++++++
 3 files changed, 317 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/generic-gpio-mux.c
drivers/gpu/drm/bridge/Kconfig +0 -0
index 1f3fc6bec842..4734f6993858 100644
@@ -54,6 +54,16 @@ config DRM_DUMB_VGA_DAC
 	  Support for non-programmable RGB to VGA DAC bridges, such as ADI
 	  ADV7123, TI THS8134 and THS8135 or passive resistor ladder DACs.

drivers/gpu/drm/bridge/Kconfig config DRM_DUMB_VGA_DAC · line 65

Quoted 10 lines
drivers/gpu/drm/bridge/Kconfig +9 -0
 
+config DRM_GENERIC_GPIO_MUX
+	tristate "Generic GPIO-controlled mux"
+	depends on OF
+	select DRM_KMS_HELPER
+	---help---
+	  This bridge driver models a GPIO-controlled display mux with one
+	  input, 2 outputs (e.g. an HDMI mux). The hardware decides which output
+	  is active, reports it as a GPIO, and the driver redirects calls to the
+	  appropriate downstream bridge (if any).

My understanding of the issue was that the mux was controllable by a GPIO, not that the GPIO would report its status. This changes a few things. How is the mux controlled then ?

Quoted 21 lines
drivers/gpu/drm/bridge/Kconfig +0 -0
 config DRM_LVDS_ENCODER
 	tristate "Transparent parallel to LVDS encoder support"
 	depends on OF
drivers/gpu/drm/bridge/Makefile +1 -0
index 7a1e0ec032e6..1c0c92667ac4 100644
@@ -3,6 +3,7 @@ obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
+obj-$(CONFIG_DRM_GENERIC_GPIO_MUX) += generic-gpio-mux.o
 obj-$(CONFIG_DRM_LVDS_ENCODER) += lvds-encoder.o
 obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o
 obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +0 -0
new file mode 100644
index 000000000000..ba08321dcc17

drivers/gpu/drm/bridge/generic-gpio-mux.c line 8

Quoted 8 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +8 -0
@@ -0,0 +1,306 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Generic gpio mux bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+

One blank line is enough.

drivers/gpu/drm/bridge/generic-gpio-mux.c line 15

Quoted 7 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +7 -0
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/of_graph.h>

Could you please sort these headers alphabetically ?

Quoted 14 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +14 -0
+#include <drm/drm_bridge.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
+
+struct gpio_display_mux {
+	struct device *dev;
+
+	struct gpio_desc *gpiod_detect;
+	int detect_irq;
+
+	struct drm_bridge bridge;
+
+	struct drm_bridge *next[2];
+};

drivers/gpu/drm/bridge/generic-gpio-mux.c line 39

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+
+static inline struct gpio_display_mux *bridge_to_gpio_display_mux(
+		struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct gpio_display_mux, bridge);
+}
+
+static irqreturn_t gpio_display_mux_det_threaded_handler(int unused, void *data)
+{
+	struct gpio_display_mux *gpio_display_mux = data;

gpio_display_mux is a long variable name. You can shorten it to mux here and below.

Quoted 6 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +6 -0
+	int active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+
+	dev_dbg(gpio_display_mux->dev, "Interrupt %d!\n", active);
+
+	if (gpio_display_mux->bridge.dev)
+		drm_kms_helper_hotplug_event(gpio_display_mux->bridge.dev);

drivers/gpu/drm/bridge/generic-gpio-mux.c line 55

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+
+	return IRQ_HANDLED;
+}
+
+static int gpio_display_mux_attach(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+			bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;

i never takes negative values, you can make it an unsigned int.

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next)
+			next->encoder = bridge->encoder;
+	}
+
+	return 0;
+}
+

drivers/gpu/drm/bridge/generic-gpio-mux.c line 75

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+static bool gpio_display_mux_mode_fixup(struct drm_bridge *bridge,
+				const struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	int active;
+	struct drm_bridge *next;
+
+	active = gpiod_get_value(gpio_display_mux->gpiod_detect);

What if the value of the GPIO changes between, let's say, this operation and gpio_display_mux_mode_set() ? This doesn't seem very stable to me. DRM/KMS hasn't been designed to have the output routing configured externally without any control from the drivers.

drivers/gpu/drm/bridge/generic-gpio-mux.c line 76

Quoted 1 line
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +1 -0
+	next = gpio_display_mux->next[active];

This will crash if gpiod_get_value() returns an error. Same for the other functions below.

Quoted 14 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +14 -0
+
+	if (next && next->funcs->mode_fixup)
+		return next->funcs->mode_fixup(next, mode, adjusted_mode);
+	else
+		return true;
+}
+
+static void gpio_display_mux_mode_set(struct drm_bridge *bridge,
+				struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	int active;

drivers/gpu/drm/bridge/generic-gpio-mux.c line 100

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+	struct drm_bridge *next;
+
+	active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+	next = gpio_display_mux->next[active];
+
+	if (next && next->funcs->mode_set)
+		next->funcs->mode_set(next, mode, adjusted_mode);
+}
+
+/**

This isn't kerneldoc, the comment should start with /*. Same comment below.

drivers/gpu/drm/bridge/generic-gpio-mux.c line 102

Quoted 2 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +2 -0
+ * Since this driver _reacts_ to mux changes, we need to make sure all
+ * downstream bridges are pre-enabled.

I'm afraid the problem scope seems bigger than I initially anticipated :-( We're in the hack territory here, and I think we need to search for a proper solution. We need to start with a detailed description of the hardware and the use cases.

Quoted 130 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +130 -0
+ */
+static void gpio_display_mux_pre_enable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->pre_enable)
+			next->funcs->pre_enable(next);
+	}
+}
+
+static void gpio_display_mux_post_disable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->post_disable)
+			next->funcs->post_disable(next);
+	}
+}
+
+/**
+ * In an ideal mux driver, only the currently selected bridge should be enabled.
+ * For the sake of simplicity, we just just enable/disable all downstream
+ * bridges at the same time.
+ */
+static void gpio_display_mux_enable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->enable)
+			next->funcs->enable(next);
+	}
+}
+
+static void gpio_display_mux_disable(struct drm_bridge *bridge)
+{
+	struct gpio_display_mux *gpio_display_mux =
+		bridge_to_gpio_display_mux(bridge);
+	struct drm_bridge *next;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+		next = gpio_display_mux->next[i];
+		if (next && next->funcs->disable)
+			next->funcs->disable(next);
+	}
+}
+
+static const struct drm_bridge_funcs gpio_display_mux_bridge_funcs = {
+	.attach = gpio_display_mux_attach,
+	.mode_fixup = gpio_display_mux_mode_fixup,
+	.disable = gpio_display_mux_disable,
+	.post_disable = gpio_display_mux_post_disable,
+	.mode_set = gpio_display_mux_mode_set,
+	.pre_enable = gpio_display_mux_pre_enable,
+	.enable = gpio_display_mux_enable,
+};
+
+static int gpio_display_mux_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct gpio_display_mux *gpio_display_mux;
+	struct device_node *port, *ep, *remote;
+	int ret;
+	u32 reg;
+
+	gpio_display_mux = devm_kzalloc(dev, sizeof(*gpio_display_mux),
+					GFP_KERNEL);
+	if (!gpio_display_mux)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, gpio_display_mux);
+	gpio_display_mux->dev = &pdev->dev;
+
+	gpio_display_mux->bridge.of_node = dev->of_node;
+
+	gpio_display_mux->gpiod_detect =
+		devm_gpiod_get(dev, "detect", GPIOD_IN);
+	if (IS_ERR(gpio_display_mux->gpiod_detect))
+		return PTR_ERR(gpio_display_mux->gpiod_detect);
+
+	gpio_display_mux->detect_irq =
+		gpiod_to_irq(gpio_display_mux->gpiod_detect);
+	if (gpio_display_mux->detect_irq < 0) {
+		dev_err(dev, "Failed to get output irq %d\n",
+			gpio_display_mux->detect_irq);
+		return -ENODEV;
+	}
+
+	port = of_graph_get_port_by_id(dev->of_node, 1);
+	if (!port) {
+		dev_err(dev, "Missing output port node\n");
+		return -EINVAL;
+	}
+
+	for_each_child_of_node(port, ep) {
+		if (!ep->name || (of_node_cmp(ep->name, "endpoint") != 0)) {
+			of_node_put(ep);
+			continue;
+		}
+
+		if (of_property_read_u32(ep, "reg", &reg) < 0 ||
+				reg >= ARRAY_SIZE(gpio_display_mux->next)) {
+			dev_err(dev,
+			    "Missing/invalid reg property for endpoint %s\n",
+				ep->full_name);
+			of_node_put(ep);
+			of_node_put(port);
+			return -EINVAL;
+		}
+
+		remote = of_graph_get_remote_port_parent(ep);
+		if (!remote) {
+			dev_err(dev,
+			    "Missing connector/bridge node for endpoint %s\n",
+				ep->full_name);

drivers/gpu/drm/bridge/generic-gpio-mux.c line 242

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+			of_node_put(ep);
+			of_node_put(port);
+			return -EINVAL;
+		}
+		of_node_put(ep);
+
+		if (of_device_is_compatible(remote, "hdmi-connector")) {
+			of_node_put(remote);
+			continue;
+		}

This special case makes me think that something is wrong. I believe the connector driver from
https://patchwork.freedesktop.org/patch/344477/?series=63328&rev=59
could help.

drivers/gpu/drm/bridge/generic-gpio-mux.c line 244

Quoted 2 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +2 -0
+
+		gpio_display_mux->next[reg] = of_drm_find_bridge(remote);

What if the connected device is a panel and not a bridge ?

Quoted 29 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +29 -0
+		if (!gpio_display_mux->next[reg]) {
+			dev_err(dev, "Waiting for external bridge %s\n",
+				remote->name);
+			of_node_put(remote);
+			of_node_put(port);
+			return -EPROBE_DEFER;
+		}
+
+		of_node_put(remote);
+	}
+	of_node_put(port);
+
+	gpio_display_mux->bridge.funcs = &gpio_display_mux_bridge_funcs;
+	drm_bridge_add(&gpio_display_mux->bridge);
+
+	ret = devm_request_threaded_irq(dev, gpio_display_mux->detect_irq,
+				NULL,
+				gpio_display_mux_det_threaded_handler,
+				IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING |
+					IRQF_ONESHOT,
+				"gpio-display-mux-det", gpio_display_mux);
+	if (ret) {
+		dev_err(dev, "Failed to request MUX_DET threaded irq\n");
+		goto err_bridge_remove;
+	}
+
+	return 0;
+
+err_bridge_remove:

drivers/gpu/drm/bridge/generic-gpio-mux.c line 283

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+	drm_bridge_remove(&gpio_display_mux->bridge);
+
+	return ret;
+}
+
+static int gpio_display_mux_remove(struct platform_device *pdev)
+{
+	struct gpio_display_mux *gpio_display_mux = platform_get_drvdata(pdev);
+
+	drm_bridge_remove(&gpio_display_mux->bridge);

If the GPIO IRQ is triggered here you'll have trouble. You need to disable the IRQ, or free it, before removing the bridge.

Quoted 23 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +23 -0
+
+	return 0;
+}
+
+static const struct of_device_id gpio_display_mux_match[] = {
+	{ .compatible = "gpio-display-mux", },
+	{},
+};
+
+struct platform_driver gpio_display_mux_driver = {
+	.probe = gpio_display_mux_probe,
+	.remove = gpio_display_mux_remove,
+	.driver = {
+		.name = "gpio-display-mux",
+		.of_match_table = gpio_display_mux_match,
+	},
+};
+
+module_platform_driver(gpio_display_mux_driver);
+
+MODULE_DESCRIPTION("GPIO-controlled display mux");
+MODULE_AUTHOR("Nicolas Boichat <[email protected]>");
+MODULE_LICENSE("GPL v2");
Signature · 7 lines
Regards,

Laurent Pinchart
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
#24 lore raw

Re: [PATCH RESEND 4/4] drm: bridge: Generic GPIO mux driver

Quoted 28 lines · Laurent Pinchart

Hi Hsin-Yi and Nicolas,

Thank you for the patch.

Quoted 22 lines · Hsin-Yi Wang

From: Nicolas Boichat <[email protected]>

This driver supports single input, 2 output display mux (e.g. HDMI mux), that provide its status via a GPIO.

---
 drivers/gpu/drm/bridge/Kconfig            |  10 +
 drivers/gpu/drm/bridge/Makefile           |   1 +
 drivers/gpu/drm/bridge/generic-gpio-mux.c | 306 ++++++++++++++++++++++
 3 files changed, 317 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/generic-gpio-mux.c
drivers/gpu/drm/bridge/Kconfig +0 -0
index 1f3fc6bec842..4734f6993858 100644
@@ -54,6 +54,16 @@ config DRM_DUMB_VGA_DAC
        Support for non-programmable RGB to VGA DAC bridges, such as ADI
        ADV7123, TI THS8134 and THS8135 or passive resistor ladder DACs.
 
Quoted 14 lines

drivers/gpu/drm/bridge/Kconfig config DRM_DUMB_VGA_DAC · line 65

Quoted 9 lines
drivers/gpu/drm/bridge/Kconfig +9 -0
+config DRM_GENERIC_GPIO_MUX
+     tristate "Generic GPIO-controlled mux"
+     depends on OF
+     select DRM_KMS_HELPER
+     ---help---
+       This bridge driver models a GPIO-controlled display mux with one
+       input, 2 outputs (e.g. an HDMI mux). The hardware decides which output
+       is active, reports it as a GPIO, and the driver redirects calls to the
+       appropriate downstream bridge (if any).

My understanding of the issue was that the mux was controllable by a GPIO, not that the GPIO would report its status. This changes a few things. How is the mux controlled then ?

There's a hardware that would control the gpio to active if HDMI is plugged. The mux driver registered a irq for this gpio, and handle cases depending on gpio status.

Quoted 106 lines
Quoted 21 lines
drivers/gpu/drm/bridge/Kconfig +0 -0
 config DRM_LVDS_ENCODER
      tristate "Transparent parallel to LVDS encoder support"
      depends on OF
drivers/gpu/drm/bridge/Makefile +1 -0
index 7a1e0ec032e6..1c0c92667ac4 100644
@@ -3,6 +3,7 @@ obj-$(CONFIG_DRM_ANALOGIX_ANX7688) += analogix-anx7688.o
 obj-$(CONFIG_DRM_ANALOGIX_ANX78XX) += analogix-anx78xx.o
 obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
 obj-$(CONFIG_DRM_DUMB_VGA_DAC) += dumb-vga-dac.o
+obj-$(CONFIG_DRM_GENERIC_GPIO_MUX) += generic-gpio-mux.o
 obj-$(CONFIG_DRM_LVDS_ENCODER) += lvds-encoder.o
 obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o
 obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +0 -0
new file mode 100644
index 000000000000..ba08321dcc17

drivers/gpu/drm/bridge/generic-gpio-mux.c line 8

Quoted 8 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +8 -0
@@ -0,0 +1,306 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Generic gpio mux bridge driver
+ *
+ * Copyright 2016 Google LLC
+ */
+
+

One blank line is enough.

drivers/gpu/drm/bridge/generic-gpio-mux.c line 15

Quoted 7 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +7 -0
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/of_graph.h>

Could you please sort these headers alphabetically ?

Quoted 14 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +14 -0
+#include <drm/drm_bridge.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_probe_helper.h>
+
+struct gpio_display_mux {
+     struct device *dev;
+
+     struct gpio_desc *gpiod_detect;
+     int detect_irq;
+
+     struct drm_bridge bridge;
+
+     struct drm_bridge *next[2];
+};

drivers/gpu/drm/bridge/generic-gpio-mux.c line 39

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+
+static inline struct gpio_display_mux *bridge_to_gpio_display_mux(
+             struct drm_bridge *bridge)
+{
+     return container_of(bridge, struct gpio_display_mux, bridge);
+}
+
+static irqreturn_t gpio_display_mux_det_threaded_handler(int unused, void *data)
+{
+     struct gpio_display_mux *gpio_display_mux = data;

gpio_display_mux is a long variable name. You can shorten it to mux here and below.

Quoted 6 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +6 -0
+     int active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+
+     dev_dbg(gpio_display_mux->dev, "Interrupt %d!\n", active);
+
+     if (gpio_display_mux->bridge.dev)
+             drm_kms_helper_hotplug_event(gpio_display_mux->bridge.dev);

drivers/gpu/drm/bridge/generic-gpio-mux.c line 55

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+
+     return IRQ_HANDLED;
+}
+
+static int gpio_display_mux_attach(struct drm_bridge *bridge)
+{
+     struct gpio_display_mux *gpio_display_mux =
+                     bridge_to_gpio_display_mux(bridge);
+     struct drm_bridge *next;
+     int i;

i never takes negative values, you can make it an unsigned int.

Quoted 17 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +17 -0
+
+     for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+             next = gpio_display_mux->next[i];
+             if (next)
+                     next->encoder = bridge->encoder;
+     }
+
+     return 0;
+}
+
+static bool gpio_display_mux_mode_fixup(struct drm_bridge *bridge,
+                             const struct drm_display_mode *mode,
+                             struct drm_display_mode *adjusted_mode)
+{
+     struct gpio_display_mux *gpio_display_mux =
+             bridge_to_gpio_display_mux(bridge);
+     int active;
Quoted 9 lines

drivers/gpu/drm/bridge/generic-gpio-mux.c line 75

Quoted 3 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +3 -0
+     struct drm_bridge *next;
+
+     active = gpiod_get_value(gpio_display_mux->gpiod_detect);

What if the value of the GPIO changes between, let's say, this operation and gpio_display_mux_mode_set() ? This doesn't seem very stable to me. DRM/KMS hasn't been designed to have the output routing configured externally without any control from the drivers.

We will fix it to store the gpio status in struct instead of reading in each bridge functions. Only in irq handler function the gpio status will be updated.

Quoted 5 lines

drivers/gpu/drm/bridge/generic-gpio-mux.c line 76

Quoted 1 line
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +1 -0
+     next = gpio_display_mux->next[active];

This will crash if gpiod_get_value() returns an error. Same for the other functions below.

Will add check for this in irq handler function.

Quoted 28 lines
Quoted 14 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +14 -0
+
+     if (next && next->funcs->mode_fixup)
+             return next->funcs->mode_fixup(next, mode, adjusted_mode);
+     else
+             return true;
+}
+
+static void gpio_display_mux_mode_set(struct drm_bridge *bridge,
+                             struct drm_display_mode *mode,
+                             struct drm_display_mode *adjusted_mode)
+{
+     struct gpio_display_mux *gpio_display_mux =
+             bridge_to_gpio_display_mux(bridge);
+     int active;

drivers/gpu/drm/bridge/generic-gpio-mux.c line 100

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+     struct drm_bridge *next;
+
+     active = gpiod_get_value(gpio_display_mux->gpiod_detect);
+     next = gpio_display_mux->next[active];
+
+     if (next && next->funcs->mode_set)
+             next->funcs->mode_set(next, mode, adjusted_mode);
+}
+
+/**

This isn't kerneldoc, the comment should start with /*. Same comment below.

Quoted 8 lines

drivers/gpu/drm/bridge/generic-gpio-mux.c line 102

Quoted 2 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +2 -0
+ * Since this driver _reacts_ to mux changes, we need to make sure all
+ * downstream bridges are pre-enabled.

I'm afraid the problem scope seems bigger than I initially anticipated :-( We're in the hack territory here, and I think we need to search for a proper solution. We need to start with a detailed description of the hardware and the use cases.

We are considering making this mux driver mt8173 oak board specific, since it seems that only this board have such design and we only have this board for testing. If so, we also only need mode_fixup bridge function. And the endpoint is anx7688 and hdmi connector. What do you think?

Quoted 137 lines
Quoted 137 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +137 -0
+ */
+static void gpio_display_mux_pre_enable(struct drm_bridge *bridge)
+{
+     struct gpio_display_mux *gpio_display_mux =
+             bridge_to_gpio_display_mux(bridge);
+     struct drm_bridge *next;
+     int i;
+
+     for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+             next = gpio_display_mux->next[i];
+             if (next && next->funcs->pre_enable)
+                     next->funcs->pre_enable(next);
+     }
+}
+
+static void gpio_display_mux_post_disable(struct drm_bridge *bridge)
+{
+     struct gpio_display_mux *gpio_display_mux =
+             bridge_to_gpio_display_mux(bridge);
+     struct drm_bridge *next;
+     int i;
+
+     for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+             next = gpio_display_mux->next[i];
+             if (next && next->funcs->post_disable)
+                     next->funcs->post_disable(next);
+     }
+}
+
+/**
+ * In an ideal mux driver, only the currently selected bridge should be enabled.
+ * For the sake of simplicity, we just just enable/disable all downstream
+ * bridges at the same time.
+ */
+static void gpio_display_mux_enable(struct drm_bridge *bridge)
+{
+     struct gpio_display_mux *gpio_display_mux =
+             bridge_to_gpio_display_mux(bridge);
+     struct drm_bridge *next;
+     int i;
+
+     for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+             next = gpio_display_mux->next[i];
+             if (next && next->funcs->enable)
+                     next->funcs->enable(next);
+     }
+}
+
+static void gpio_display_mux_disable(struct drm_bridge *bridge)
+{
+     struct gpio_display_mux *gpio_display_mux =
+             bridge_to_gpio_display_mux(bridge);
+     struct drm_bridge *next;
+     int i;
+
+     for (i = 0; i < ARRAY_SIZE(gpio_display_mux->next); i++) {
+             next = gpio_display_mux->next[i];
+             if (next && next->funcs->disable)
+                     next->funcs->disable(next);
+     }
+}
+
+static const struct drm_bridge_funcs gpio_display_mux_bridge_funcs = {
+     .attach = gpio_display_mux_attach,
+     .mode_fixup = gpio_display_mux_mode_fixup,
+     .disable = gpio_display_mux_disable,
+     .post_disable = gpio_display_mux_post_disable,
+     .mode_set = gpio_display_mux_mode_set,
+     .pre_enable = gpio_display_mux_pre_enable,
+     .enable = gpio_display_mux_enable,
+};
+
+static int gpio_display_mux_probe(struct platform_device *pdev)
+{
+     struct device *dev = &pdev->dev;
+     struct gpio_display_mux *gpio_display_mux;
+     struct device_node *port, *ep, *remote;
+     int ret;
+     u32 reg;
+
+     gpio_display_mux = devm_kzalloc(dev, sizeof(*gpio_display_mux),
+                                     GFP_KERNEL);
+     if (!gpio_display_mux)
+             return -ENOMEM;
+
+     platform_set_drvdata(pdev, gpio_display_mux);
+     gpio_display_mux->dev = &pdev->dev;
+
+     gpio_display_mux->bridge.of_node = dev->of_node;
+
+     gpio_display_mux->gpiod_detect =
+             devm_gpiod_get(dev, "detect", GPIOD_IN);
+     if (IS_ERR(gpio_display_mux->gpiod_detect))
+             return PTR_ERR(gpio_display_mux->gpiod_detect);
+
+     gpio_display_mux->detect_irq =
+             gpiod_to_irq(gpio_display_mux->gpiod_detect);
+     if (gpio_display_mux->detect_irq < 0) {
+             dev_err(dev, "Failed to get output irq %d\n",
+                     gpio_display_mux->detect_irq);
+             return -ENODEV;
+     }
+
+     port = of_graph_get_port_by_id(dev->of_node, 1);
+     if (!port) {
+             dev_err(dev, "Missing output port node\n");
+             return -EINVAL;
+     }
+
+     for_each_child_of_node(port, ep) {
+             if (!ep->name || (of_node_cmp(ep->name, "endpoint") != 0)) {
+                     of_node_put(ep);
+                     continue;
+             }
+
+             if (of_property_read_u32(ep, "reg", &reg) < 0 ||
+                             reg >= ARRAY_SIZE(gpio_display_mux->next)) {
+                     dev_err(dev,
+                         "Missing/invalid reg property for endpoint %s\n",
+                             ep->full_name);
+                     of_node_put(ep);
+                     of_node_put(port);
+                     return -EINVAL;
+             }
+
+             remote = of_graph_get_remote_port_parent(ep);
+             if (!remote) {
+                     dev_err(dev,
+                         "Missing connector/bridge node for endpoint %s\n",
+                             ep->full_name);
+                     of_node_put(ep);
+                     of_node_put(port);
+                     return -EINVAL;
+             }
+             of_node_put(ep);
+
+             if (of_device_is_compatible(remote, "hdmi-connector")) {
Quoted 9 lines

drivers/gpu/drm/bridge/generic-gpio-mux.c line 242

Quoted 3 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +3 -0
+                     of_node_put(remote);
+                     continue;
+             }

This special case makes me think that something is wrong. I believe the connector driver from
https://patchwork.freedesktop.org/patch/344477/?series=63328&rev=59
could help.

Thanks for pointing this, we can remove the special case (hdmi) handling here.

Quoted 5 lines

drivers/gpu/drm/bridge/generic-gpio-mux.c line 244

Quoted 2 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +2 -0
+
+             gpio_display_mux->next[reg] = of_drm_find_bridge(remote);

What if the connected device is a panel and not a bridge ?

If this is oak specific, then we don't need to handle panel case.

Quoted 34 lines
Quoted 34 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +34 -0
+             if (!gpio_display_mux->next[reg]) {
+                     dev_err(dev, "Waiting for external bridge %s\n",
+                             remote->name);
+                     of_node_put(remote);
+                     of_node_put(port);
+                     return -EPROBE_DEFER;
+             }
+
+             of_node_put(remote);
+     }
+     of_node_put(port);
+
+     gpio_display_mux->bridge.funcs = &gpio_display_mux_bridge_funcs;
+     drm_bridge_add(&gpio_display_mux->bridge);
+
+     ret = devm_request_threaded_irq(dev, gpio_display_mux->detect_irq,
+                             NULL,
+                             gpio_display_mux_det_threaded_handler,
+                             IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING |
+                                     IRQF_ONESHOT,
+                             "gpio-display-mux-det", gpio_display_mux);
+     if (ret) {
+             dev_err(dev, "Failed to request MUX_DET threaded irq\n");
+             goto err_bridge_remove;
+     }
+
+     return 0;
+
+err_bridge_remove:
+     drm_bridge_remove(&gpio_display_mux->bridge);
+
+     return ret;
+}
+
Quoted 9 lines

drivers/gpu/drm/bridge/generic-gpio-mux.c line 283

Quoted 5 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +5 -0
+static int gpio_display_mux_remove(struct platform_device *pdev)
+{
+     struct gpio_display_mux *gpio_display_mux = platform_get_drvdata(pdev);
+
+     drm_bridge_remove(&gpio_display_mux->bridge);

If the GPIO IRQ is triggered here you'll have trouble. You need to disable the IRQ, or free it, before removing the bridge.

Will fix this. Thanks

Quoted 28 lines
Quoted 13 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +13 -0
+
+     return 0;
+}
+
+static const struct of_device_id gpio_display_mux_match[] = {
+     { .compatible = "gpio-display-mux", },
+     {},
+};
+
+struct platform_driver gpio_display_mux_driver = {
+     .probe = gpio_display_mux_probe,
+     .remove = gpio_display_mux_remove,
+     .driver = {

drivers/gpu/drm/bridge/generic-gpio-mux.c line 306

Quoted 10 lines
drivers/gpu/drm/bridge/generic-gpio-mux.c new file +10 -0
+             .name = "gpio-display-mux",
+             .of_match_table = gpio_display_mux_match,
+     },
+};
+
+module_platform_driver(gpio_display_mux_driver);
+
+MODULE_DESCRIPTION("GPIO-controlled display mux");
+MODULE_AUTHOR("Nicolas Boichat <[email protected]>");
+MODULE_LICENSE("GPL v2");

--
Regards,

Laurent Pinchart