Re: [PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Simon Horman]

On Fri, Jan 12, 2018 at 02:58:48AM +0200, Laurent Pinchart wrote:
> Hello,
> 
> This patch series addresses a design mistake that dates back from the initial
> DU support. Support for the LVDS encoders, which are IP cores separate from
> the DU, was bundled in the DU driver. Worse, both the DU and LVDS were
> described through a single DT node.
> 
> To fix the, patches 01/10 and 02/10 define new DT bindings for the LVDS
> encoders, and deprecate their description inside the DU bindings. To retain
> backward compatibility with existing DT, patch 03/10 then patches the device
> tree at runtime to convert the legacy bindings to the new ones.
> 
> With the DT side addressed, patch 04/10 then converts the LVDS support code to
> a separate bridge driver. After a small fix to the Porter board device tree in
> patch 05/10, patches 06/10 to 10/10 then update all the device tree sources to
> the new LVDS encoders bindings.
> 
> I decided to go for live DT patching in patch 03/10 because implementing
> support for both the legacy and new bindings in the driver would have been
> very intrusive, and prevented further cleanups. I'm in a way both proud and 
> ashamed of that patch that I would call a neat and dirty hack. It was fun to
> write perhaps in a similar way that one would enjoy researching and developing
> proof-of-concepts for security holes: they're good and bad at the same time.
> 
> Anyway, with this philosophical considerations aside, there were a few
> shortcomings in the OF API that I worked around with local code in the driver.
> If anyone is interested in performing similar live DT patching I think we
> could move some of the code to the OF core. For instance I was surprised
> to not find a device node lookup by path function that would start at a
> given node instead of the root of the live device tree, and had to write
> rcar_du_of_find_node_by_path(). Utility functions to add or modify properties
> or to rename nodes could similarly be added.
> 
> Laurent Pinchart (10):
>   dt-bindings: display: renesas: Add R-Car LVDS encoder DT bindings
>   dt-bindings: display: renesas: Deprecate LVDS support in the DU
> bindings
>   drm: rcar-du: Fix legacy DT to create LVDS encoder nodes
>   drm: rcar-du: Convert LVDS encoder code to bridge driver
>   ARM: dts: porter: Fix HDMI output routing
>   ARM: dts: r8a7790: Convert to new LVDS DT bindings
>   ARM: dts: r8a7791: Convert to new LVDS DT bindings
>   ARM: dts: r8a7793: Convert to new LVDS DT bindings
>   arm64: dts: renesas: r8a7795: Convert to new LVDS DT bindings
>   arm64: dts: renesas: r8a7796: Convert to new LVDS DT bindings

I understand that there will be a v2 to address review of the non DTS
patches. I am marking the DTS patches as "Changes Requested" as they
are dependent on the bindings patches from my PoV.

___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Simon Horman]

On Fri, Jan 12, 2018 at 03:48:25PM +0200, Laurent Pinchart wrote:
> Hi Geert,
> 
> On Friday, 12 January 2018 11:47:03 EET Geert Uytterhoeven wrote:
> > On Fri, Jan 12, 2018 at 1:58 AM, Laurent Pinchart wrote:
> > > This patch series addresses a design mistake that dates back from the
> > > initial DU support. Support for the LVDS encoders, which are IP cores
> > > separate from the DU, was bundled in the DU driver. Worse, both the DU
> > > and LVDS were described through a single DT node.
> > > 
> > > To fix the, patches 01/10 and 02/10 define new DT bindings for the LVDS
> > > encoders, and deprecate their description inside the DU bindings. To
> > > retain backward compatibility with existing DT, patch 03/10 then patches
> > > the device tree at runtime to convert the legacy bindings to the new ones.
> > 
> > Looks like we will have to postpone the R-Car Gen2 Modern DT flag day
> > again by a few kernel releases?
> 
> Why so ? We don't have to drop support for all legacy DT bindings at the same 
> time, do we ? We can switch to the new-style clock bindings on Gen2 already, 
> and drop the legacy LVDS bindings later.

To clarify, after this patchset.
* Old DTs work with old and new kernels.
* New DTs require new kernels.

___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Simon Horman]

On Mon, Jan 15, 2018 at 09:00:59AM +0200, Laurent Pinchart wrote:
> Hi Simon,
> 
> On Monday, 15 January 2018 08:57:48 EET Simon Horman wrote:
> > On Fri, Jan 12, 2018 at 03:48:25PM +0200, Laurent Pinchart wrote:
> > > On Friday, 12 January 2018 11:47:03 EET Geert Uytterhoeven wrote:
> > >> On Fri, Jan 12, 2018 at 1:58 AM, Laurent Pinchart wrote:
> > >>> This patch series addresses a design mistake that dates back from the
> > >>> initial DU support. Support for the LVDS encoders, which are IP cores
> > >>> separate from the DU, was bundled in the DU driver. Worse, both the DU
> > >>> and LVDS were described through a single DT node.
> > >>> 
> > >>> To fix the, patches 01/10 and 02/10 define new DT bindings for the
> > >>> LVDS encoders, and deprecate their description inside the DU bindings.
> > >>> To retain backward compatibility with existing DT, patch 03/10 then
> > >>> patches the device tree at runtime to convert the legacy bindings to the
> > >>> new ones.
> > >> 
> > >> Looks like we will have to postpone the R-Car Gen2 Modern DT flag day
> > >> again by a few kernel releases?
> > > 
> > > Why so ? We don't have to drop support for all legacy DT bindings at the
> > > same time, do we ? We can switch to the new-style clock bindings on Gen2
> > > already, and drop the legacy LVDS bindings later.
> > 
> > To clarify, after this patchset.
> > * Old DTs work with old and new kernels.
> > * New DTs require new kernels.
> 
> That's correct. I've tested old DTs with new kernels successfully on Lager, 
> Salvator-X (both H3 ES1.x and M3-W) and Salvator-XS.

Thanks, no objections to this approach from my side.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Laurent Pinchart]

Hi Simon,

On Monday, 15 January 2018 08:57:48 EET Simon Horman wrote:
> On Fri, Jan 12, 2018 at 03:48:25PM +0200, Laurent Pinchart wrote:
> > On Friday, 12 January 2018 11:47:03 EET Geert Uytterhoeven wrote:
> >> On Fri, Jan 12, 2018 at 1:58 AM, Laurent Pinchart wrote:
> >>> This patch series addresses a design mistake that dates back from the
> >>> initial DU support. Support for the LVDS encoders, which are IP cores
> >>> separate from the DU, was bundled in the DU driver. Worse, both the DU
> >>> and LVDS were described through a single DT node.
> >>> 
> >>> To fix the, patches 01/10 and 02/10 define new DT bindings for the
> >>> LVDS encoders, and deprecate their description inside the DU bindings.
> >>> To retain backward compatibility with existing DT, patch 03/10 then
> >>> patches the device tree at runtime to convert the legacy bindings to the
> >>> new ones.
> >> 
> >> Looks like we will have to postpone the R-Car Gen2 Modern DT flag day
> >> again by a few kernel releases?
> > 
> > Why so ? We don't have to drop support for all legacy DT bindings at the
> > same time, do we ? We can switch to the new-style clock bindings on Gen2
> > already, and drop the legacy LVDS bindings later.
> 
> To clarify, after this patchset.
> * Old DTs work with old and new kernels.
> * New DTs require new kernels.

That's correct. I've tested old DTs with new kernels successfully on Lager, 
Salvator-X (both H3 ES1.x and M3-W) and Salvator-XS.

-- 
Regards,

Laurent Pinchart

___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Laurent Pinchart]

Hi Geert,

On Friday, 12 January 2018 11:47:03 EET Geert Uytterhoeven wrote:
> On Fri, Jan 12, 2018 at 1:58 AM, Laurent Pinchart wrote:
> > This patch series addresses a design mistake that dates back from the
> > initial DU support. Support for the LVDS encoders, which are IP cores
> > separate from the DU, was bundled in the DU driver. Worse, both the DU
> > and LVDS were described through a single DT node.
> > 
> > To fix the, patches 01/10 and 02/10 define new DT bindings for the LVDS
> > encoders, and deprecate their description inside the DU bindings. To
> > retain backward compatibility with existing DT, patch 03/10 then patches
> > the device tree at runtime to convert the legacy bindings to the new ones.
> 
> Looks like we will have to postpone the R-Car Gen2 Modern DT flag day
> again by a few kernel releases?

Why so ? We don't have to drop support for all legacy DT bindings at the same 
time, do we ? We can switch to the new-style clock bindings on Gen2 already, 
and drop the legacy LVDS bindings later.

-- 
Regards,

Laurent Pinchart

___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Geert Uytterhoeven]

Hi Laurent,

On Fri, Jan 12, 2018 at 1:58 AM, Laurent Pinchart
 wrote:
> This patch series addresses a design mistake that dates back from the initial
> DU support. Support for the LVDS encoders, which are IP cores separate from
> the DU, was bundled in the DU driver. Worse, both the DU and LVDS were
> described through a single DT node.
>
> To fix the, patches 01/10 and 02/10 define new DT bindings for the LVDS
> encoders, and deprecate their description inside the DU bindings. To retain
> backward compatibility with existing DT, patch 03/10 then patches the device
> tree at runtime to convert the legacy bindings to the new ones.

Looks like we will have to postpone the R-Car Gen2 Modern DT flag day
again by a few kernel releases?

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[PATCH 00/10] R-Car DU: Convert LVDS code to bridge driver

[Laurent Pinchart]

Hello,

This patch series addresses a design mistake that dates back from the initial
DU support. Support for the LVDS encoders, which are IP cores separate from
the DU, was bundled in the DU driver. Worse, both the DU and LVDS were
described through a single DT node.

To fix the, patches 01/10 and 02/10 define new DT bindings for the LVDS
encoders, and deprecate their description inside the DU bindings. To retain
backward compatibility with existing DT, patch 03/10 then patches the device
tree at runtime to convert the legacy bindings to the new ones.

With the DT side addressed, patch 04/10 then converts the LVDS support code to
a separate bridge driver. After a small fix to the Porter board device tree in
patch 05/10, patches 06/10 to 10/10 then update all the device tree sources to
the new LVDS encoders bindings.

I decided to go for live DT patching in patch 03/10 because implementing
support for both the legacy and new bindings in the driver would have been
very intrusive, and prevented further cleanups. I'm in a way both proud and 
ashamed of that patch that I would call a neat and dirty hack. It was fun to
write perhaps in a similar way that one would enjoy researching and developing
proof-of-concepts for security holes: they're good and bad at the same time.

Anyway, with this philosophical considerations aside, there were a few
shortcomings in the OF API that I worked around with local code in the driver.
If anyone is interested in performing similar live DT patching I think we
could move some of the code to the OF core. For instance I was surprised
to not find a device node lookup by path function that would start at a
given node instead of the root of the live device tree, and had to write
rcar_du_of_find_node_by_path(). Utility functions to add or modify properties
or to rename nodes could similarly be added.

Laurent Pinchart (10):
  dt-bindings: display: renesas: Add R-Car LVDS encoder DT bindings
  dt-bindings: display: renesas: Deprecate LVDS support in the DU
bindings
  drm: rcar-du: Fix legacy DT to create LVDS encoder nodes
  drm: rcar-du: Convert LVDS encoder code to bridge driver
  ARM: dts: porter: Fix HDMI output routing
  ARM: dts: r8a7790: Convert to new LVDS DT bindings
  ARM: dts: r8a7791: Convert to new LVDS DT bindings
  ARM: dts: r8a7793: Convert to new LVDS DT bindings
  arm64: dts: renesas: r8a7795: Convert to new LVDS DT bindings
  arm64: dts: renesas: r8a7796: Convert to new LVDS DT bindings

 .../bindings/display/bridge/renesas,lvds.txt   |  54 ++
 .../devicetree/bindings/display/renesas,du.txt |  26 +-
 MAINTAINERS|   1 +
 arch/arm/boot/dts/r8a7790-lager.dts|  22 +-
 arch/arm/boot/dts/r8a7790.dtsi |  61 ++-
 arch/arm/boot/dts/r8a7791-koelsch.dts  |  10 +-
 arch/arm/boot/dts/r8a7791-porter.dts   |  18 +-
 arch/arm/boot/dts/r8a7791.dtsi |  35 +-
 arch/arm/boot/dts/r8a7793-gose.dts |  10 +-
 arch/arm/boot/dts/r8a7793.dtsi |  35 +-
 .../boot/dts/renesas/r8a7795-es1-salvator-x.dts|   3 +-
 arch/arm64/boot/dts/renesas/r8a7795-h3ulcb.dts |   3 +-
 arch/arm64/boot/dts/renesas/r8a7795-salvator-x.dts |   3 +-
 .../arm64/boot/dts/renesas/r8a7795-salvator-xs.dts |   3 +-
 arch/arm64/boot/dts/renesas/r8a7795.dtsi   |  35 +-
 arch/arm64/boot/dts/renesas/r8a7796-m3ulcb.dts |   3 +-
 arch/arm64/boot/dts/renesas/r8a7796-salvator-x.dts |   3 +-
 arch/arm64/boot/dts/renesas/r8a7796.dtsi   |  35 +-
 drivers/gpu/drm/rcar-du/Kconfig|   5 +-
 drivers/gpu/drm/rcar-du/Makefile   |   6 +-
 drivers/gpu/drm/rcar-du/rcar_du_drv.c  |  21 +-
 drivers/gpu/drm/rcar-du/rcar_du_drv.h  |   5 -
 drivers/gpu/drm/rcar-du/rcar_du_encoder.c  | 175 +--
 drivers/gpu/drm/rcar-du/rcar_du_encoder.h  |  12 -
 drivers/gpu/drm/rcar-du/rcar_du_kms.c  |  14 +-
 drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c  |  93 
 drivers/gpu/drm/rcar-du/rcar_du_lvdscon.h  |  24 -
 drivers/gpu/drm/rcar-du/rcar_du_lvdsenc.c  | 270 --
 drivers/gpu/drm/rcar-du/rcar_du_lvdsenc.h  |  64 ---
 drivers/gpu/drm/rcar-du/rcar_du_of.c   | 440 
 drivers/gpu/drm/rcar-du/rcar_du_of.h   |  16 +
 drivers/gpu/drm/rcar-du/rcar_du_of_lvds.dts|  82 +++
 drivers/gpu/drm/rcar-du/rcar_lvds.c| 554 +
 33 files changed, 1420 insertions(+), 721 deletions(-)
 create mode 100644 
Documentation/devicetree/bindings/display/bridge/renesas,lvds.txt
 delete mode 100644 drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
 delete mode 100644 drivers/gpu/drm/rcar-du/rcar_du_lvdscon.h
 delete mode 100644 drivers/gpu/drm/rcar-du/rcar_du_lvdsenc.c
 delete mode 100644 drivers/gpu/drm/rcar-du/rcar_du_lvdsenc.h
 create mode 100644 drivers/gpu/drm/rcar-du/rcar_du_of.c
 create mode