[PATCH 0/2] drm/bridge: ti-sn65dsi83: Add reverse lvds lanes DT property

[PATCH 0/2] drm/bridge: ti-sn65dsi83: Add reverse lvds lanes DT property

[Wojciech Dubowik]

From: Wojciech Dubowik <Wojciech.Dubowik@mt.com>

Add lvds reverse lanes device tree property. When set, the
endpoint lvds data lanes are configured in <4 3 2 1>
order instead of default <1 2 3 4>. This is needed for layouts
where lanes have been swapped.

Wojciech Dubowik (2):
  dt-bindings: display: sn65dsi83: Add reverse lanes property
  drm/bridge: ti-sn65dsi83: Add reverse lvds lanes property

 .../devicetree/bindings/display/bridge/ti,sn65dsi83.yaml  | 4 ++++
 drivers/gpu/drm/bridge/ti-sn65dsi83.c                     | 8 ++++++++
 2 files changed, 12 insertions(+)

-- 
2.47.3

[PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[Wojciech Dubowik]

From: Wojciech Dubowik <Wojciech.Dubowik@mt.com>

Add reverse lvds lanes property for endpoints. It reverses
the data-lanes order i.e. <4 3 2 1> instead of default
<1 2 3 4>.

Signed-off-by: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
---
 .../devicetree/bindings/display/bridge/ti,sn65dsi83.yaml      | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
index e69b6343a8eb..44483e808af9 100644
--- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
@@ -115,6 +115,10 @@ $defs:
             enum: [100, 200]
             default: 200
 
+          ti,lvds-reverse-lanes:
+            description: Reverse the order of LVDS lanes.
+            type: boolean
+
           ti,lvds-vod-swing-clock-microvolt:
             description: LVDS diferential output voltage <min max> for clock
               lanes in microvolts.
-- 
2.47.3

Re: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The patch introduces a custom vendor-specific boolean property `ti,lvds-reverse-lanes` to indicate LVDS lane reversal. This is a device tree anti-pattern because the referenced standard `video-interfaces.yaml` already provides the `data-lanes` property for this exact purpose (e.g., `data-lanes = <4 3 2 1>;`).
--

--- Patch [1]: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property ---
commit c30eea01a74188461b7e527f71d282d68b885ec2
Author: Wojciech Dubowik <Wojciech.Dubowik@mt.com>

dt-bindings: display: sn65dsi83: Add reverse lanes property

This patch adds a boolean property to reverse the LVDS data lanes order for
endpoints.

> diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
[ ... ]
> @@ -115,6 +115,10 @@ $defs:
>              enum: [100, 200]
>              default: 200
>  
> +          ti,lvds-reverse-lanes:
> +            description: Reverse the order of LVDS lanes.
> +            type: boolean

[Severity: Low]
Is it necessary to introduce a custom vendor-specific boolean property
for this?

Since video-interfaces.yaml already provides the standard data-lanes property
which can be used to describe the exact lane routing, does this hardware
require a distinct property instead of using the standard data-lanes
approach with data-lanes = <4 3 2 1>?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260707070244.230546-1-wojciech.dubowik@mt.com?part=1

Re: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[Conor Dooley]

[-- Attachment #1: Type: text/plain, Size: 1926 bytes --]

On Tue, Jul 07, 2026 at 07:45:00AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Low] The patch introduces a custom vendor-specific boolean property `ti,lvds-reverse-lanes` to indicate LVDS lane reversal. This is a device tree anti-pattern because the referenced standard `video-interfaces.yaml` already provides the `data-lanes` property for this exact purpose (e.g., `data-lanes = <4 3 2 1>;`).
> --
> 
> --- Patch [1]: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property ---
> commit c30eea01a74188461b7e527f71d282d68b885ec2
> Author: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> 
> dt-bindings: display: sn65dsi83: Add reverse lanes property
> 
> This patch adds a boolean property to reverse the LVDS data lanes order for
> endpoints.
> 
> > diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> [ ... ]
> > @@ -115,6 +115,10 @@ $defs:
> >              enum: [100, 200]
> >              default: 200
> >  
> > +          ti,lvds-reverse-lanes:
> > +            description: Reverse the order of LVDS lanes.
> > +            type: boolean
> 
> [Severity: Low]
> Is it necessary to introduce a custom vendor-specific boolean property
> for this?
> 
> Since video-interfaces.yaml already provides the standard data-lanes property
> which can be used to describe the exact lane routing, does this hardware
> require a distinct property instead of using the standard data-lanes
> approach with data-lanes = <4 3 2 1>?

I don't know much about display devices, but this seems like a
reasonable thing to me. Why would it not work?

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[Wojciech Dubowik]

On Tue, Jul 07, 2026 at 05:21:19PM +0100, Conor Dooley wrote:
Hi Conor,

> On Tue, Jul 07, 2026 at 07:45:00AM +0000, sashiko-bot@kernel.org wrote:
> > Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> > - [Low] The patch introduces a custom vendor-specific boolean property `ti,lvds-reverse-lanes` to indicate LVDS lane reversal. This is a device tree anti-pattern because the referenced standard `video-interfaces.yaml` already provides the `data-lanes` property for this exact purpose (e.g., `data-lanes = <4 3 2 1>;`).
> > --
> > 
> > --- Patch [1]: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property ---
> > commit c30eea01a74188461b7e527f71d282d68b885ec2
> > Author: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> > 
> > dt-bindings: display: sn65dsi83: Add reverse lanes property
> > 
> > This patch adds a boolean property to reverse the LVDS data lanes order for
> > endpoints.
> > 
> > > diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > > --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > > +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > [ ... ]
> > > @@ -115,6 +115,10 @@ $defs:
> > >              enum: [100, 200]
> > >              default: 200
> > >  
> > > +          ti,lvds-reverse-lanes:
> > > +            description: Reverse the order of LVDS lanes.
> > > +            type: boolean
> > 
> > [Severity: Low]
> > Is it necessary to introduce a custom vendor-specific boolean property
> > for this?
> > 
> > Since video-interfaces.yaml already provides the standard data-lanes property
> > which can be used to describe the exact lane routing, does this hardware
> > require a distinct property instead of using the standard data-lanes
> > approach with data-lanes = <4 3 2 1>?
> 
> I don't know much about display devices, but this seems like a
> reasonable thing to me. Why would it not work?

It makes perfectly sense. I am working on it already and will send soon V2.

Re: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[Alexander Stein]

Hi,

Am Dienstag, 7. Juli 2026, 09:02:42 CEST schrieb Wojciech Dubowik:
> From: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> 
> Add reverse lvds lanes property for endpoints. It reverses
> the data-lanes order i.e. <4 3 2 1> instead of default
> <1 2 3 4>.

Wouldn't it be more reasonable to parse the data-lanes property and check
for the order instead of introducing a new property?

Best regards
Alexander

> 
> Signed-off-by: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> ---
>  .../devicetree/bindings/display/bridge/ti,sn65dsi83.yaml      | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> index e69b6343a8eb..44483e808af9 100644
> --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> @@ -115,6 +115,10 @@ $defs:
>              enum: [100, 200]
>              default: 200
>  
> +          ti,lvds-reverse-lanes:
> +            description: Reverse the order of LVDS lanes.
> +            type: boolean
> +
>            ti,lvds-vod-swing-clock-microvolt:
>              description: LVDS diferential output voltage <min max> for clock
>                lanes in microvolts.
> 


-- 
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
http://www.tq-group.com/

Re: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[Laurent Pinchart]

On Thu, Jul 09, 2026 at 10:33:36AM +0200, Alexander Stein wrote:
> Am Dienstag, 7. Juli 2026, 09:02:42 CEST schrieb Wojciech Dubowik:
> > From: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> > 
> > Add reverse lvds lanes property for endpoints. It reverses
> > the data-lanes order i.e. <4 3 2 1> instead of default
> > <1 2 3 4>.
> 
> Wouldn't it be more reasonable to parse the data-lanes property and check
> for the order instead of introducing a new property?

I think so. That's the intended usage of data-lanes, the property is
meant to convey the order of lanes in addition to the count. Otherwise
we would have used a data-lanes-count property instead.

> > Signed-off-by: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> > ---
> >  .../devicetree/bindings/display/bridge/ti,sn65dsi83.yaml      | 4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > index e69b6343a8eb..44483e808af9 100644
> > --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > @@ -115,6 +115,10 @@ $defs:
> >              enum: [100, 200]
> >              default: 200
> >  
> > +          ti,lvds-reverse-lanes:
> > +            description: Reverse the order of LVDS lanes.
> > +            type: boolean
> > +
> >            ti,lvds-vod-swing-clock-microvolt:
> >              description: LVDS diferential output voltage <min max> for clock
> >                lanes in microvolts.

-- 
Regards,

Laurent Pinchart

Re: [PATCH 1/2] dt-bindings: display: sn65dsi83: Add reverse lanes property

[Wojciech Dubowik]

On Thu, Jul 09, 2026 at 12:17:05PM +0300, Laurent Pinchart wrote:
> On Thu, Jul 09, 2026 at 10:33:36AM +0200, Alexander Stein wrote:
> > Am Dienstag, 7. Juli 2026, 09:02:42 CEST schrieb Wojciech Dubowik:
> > > From: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> > > 
> > > Add reverse lvds lanes property for endpoints. It reverses
> > > the data-lanes order i.e. <4 3 2 1> instead of default
> > > <1 2 3 4>.
> > 
> > Wouldn't it be more reasonable to parse the data-lanes property and check
> > for the order instead of introducing a new property?
> 
> I think so. That's the intended usage of data-lanes, the property is
> meant to convey the order of lanes in addition to the count. Otherwise
> we would have used a data-lanes-count property instead.

I will send v2 with data-lanes parsing instead of adding new dt property.
Regards,
Wojtek

> 
> > > Signed-off-by: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
> > > ---
> > >  .../devicetree/bindings/display/bridge/ti,sn65dsi83.yaml      | 4 ++++
> > >  1 file changed, 4 insertions(+)
> > > 
> > > diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > > index e69b6343a8eb..44483e808af9 100644
> > > --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > > +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml
> > > @@ -115,6 +115,10 @@ $defs:
> > >              enum: [100, 200]
> > >              default: 200
> > >  
> > > +          ti,lvds-reverse-lanes:
> > > +            description: Reverse the order of LVDS lanes.
> > > +            type: boolean
> > > +
> > >            ti,lvds-vod-swing-clock-microvolt:
> > >              description: LVDS diferential output voltage <min max> for clock
> > >                lanes in microvolts.
> 
> -- 
> Regards,
> 
> Laurent Pinchart

[PATCH 2/2] drm/bridge: ti-sn65dsi83: Add reverse lvds lanes property

[Wojciech Dubowik]

From: Wojciech Dubowik <Wojciech.Dubowik@mt.com>

Add parsing of lvds reverse lanes device tree property. When
set, the endpoint lvds data lanes are configured in <4 3 2 1>
order instead of default <1 2 3 4>. This is needed for layouts
where lanes have been swapped.

Signed-off-by: Wojciech Dubowik <Wojciech.Dubowik@mt.com>
---
 drivers/gpu/drm/bridge/ti-sn65dsi83.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
index 42b451432bbb..2ab2764065c4 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
@@ -163,6 +163,7 @@ struct sn65dsi83 {
 	struct regulator		*vcc;
 	bool				lvds_dual_link;
 	bool				lvds_dual_link_even_odd_swap;
+	bool				lvds_reverse_lanes_conf[2];
 	int				lvds_vod_swing_conf[2];
 	int				lvds_term_conf[2];
 	int				irq;
@@ -644,6 +645,10 @@ static void sn65dsi83_atomic_pre_enable(struct drm_bridge *bridge,
 	regmap_write(ctx->regmap, REG_LVDS_LANE,
 		     (ctx->lvds_dual_link_even_odd_swap ?
 		      REG_LVDS_LANE_EVEN_ODD_SWAP : 0) |
+		     (ctx->lvds_reverse_lanes_conf[CHANNEL_A] ?
+		      REG_LVDS_LANE_CHA_REVERSE_LVDS : 0) |
+		     (ctx->lvds_reverse_lanes_conf[CHANNEL_B] ?
+		      REG_LVDS_LANE_CHB_REVERSE_LVDS : 0) |
 		     (ctx->lvds_term_conf[CHANNEL_A] ?
 			  REG_LVDS_LANE_CHA_LVDS_TERM : 0) |
 		     (ctx->lvds_term_conf[CHANNEL_B] ?
@@ -854,6 +859,9 @@ static int sn65dsi83_parse_lvds_endpoint(struct sn65dsi83 *ctx, int channel)
 		goto exit;
 	}
 
+	ctx->lvds_reverse_lanes_conf[channel] =
+		of_property_read_bool(endpoint, "ti,lvds-reverse-lanes");
+
 	ret_data = of_property_read_u32_array(endpoint, "ti,lvds-vod-swing-data-microvolt",
 					lvds_vod_swing_data, ARRAY_SIZE(lvds_vod_swing_data));
 	if (ret_data != 0 && ret_data != -EINVAL) {
-- 
2.47.3