[PATCH v3 0/7] drm/vc4: Fixup DT and DT binding issues from recent patchset

[Dave Stevenson <dave.stevenson@xxxxxxxxxxxxxxx>]

I missed the DT errors from the recent patchset[1] (DT patches
in linux-next via Florian, DRM bindings patches on dri-misc-next)
as Rob's bot report got spam filtered, so this is a fixup set.

Largely it was changes to number of interrupts or clocks in the
bindings, so those are now covered.

I've fixed up the missing "interrupt-controller" flags for 2711
and 2712 whilst here.

I can't get my head around what is meant to happen with ranges:
"soc@107c000000: firmware: 'ranges' is a required property"
The meaning seems obvious.

However if I add it then I get:
"firmware: '#address-cells', '#size-cells', 'dma-ranges', 'ranges' do
not match any of the regexes: 'pinctrl-[0-9]+'
from schema $id: http://devicetree.org/schemas/arm/bcm/raspberrypi,bcm2835-firmware.yaml#

There's obviously some other flag I need to set in the bindings,
but I can't work it out. We have similar errors for all the Pi
platforms for one or more nodes.
Please advise and I'll happily fix them all.

Thanks
  Dave

[1] https://lore.kernel.org/linux-arm-kernel/20241025-drm-vc4-2712-support-v2-0-35efa83c8fc0@xxxxxxxxxxxxxxx/

Signed-off-by: Dave Stevenson <dave.stevenson@xxxxxxxxxxxxxxx>
---
Changes in v3:
- Fixed up indentation on 1/7. (I fixed it once, but obviously reworked
  things and lost it).
- Link to v2: https://lore.kernel.org/r/20241212-dt-bcm2712-fixes-v2-0-35986e04d0f4@xxxxxxxxxxxxxxx

Thanks to Stefan and Krzysztof for their reviews.
Hopefully I've addressed all points raised in the correct manner.

Changes in v2:
- Commits have now be merged from drm-misc-next to linux-next, so all
  commit hashes are valid on linux-next.
- 1/7 Removed references to "previous commit". Fixed up indentation.
  Added maxItems
- 2/7 Defined widest constraints
- 3/7 Added maxItems and removed reference to Linux
- 4/7 Described the errors. Split into two for fix of node name vs addr
  being wrong.
- Added new patch removing "required" for interrupt-controller and
  interrupt-cells for bcm2836-l1-intc
- 5/7 (now 7/7) Removed the intc node for 2712 - it's irrelevant on 64bit systems
- 6/7 dropped as updating the binding is the correct answer
- 7/7 dropped. simple-bus claims ranges is required, but adding it
  creates other errors. I'm unclear as to the right solution.

- Link to v1: https://lore.kernel.org/r/20241202-dt-bcm2712-fixes-v1-0-fac67cc2f98a@xxxxxxxxxxxxxxx

---
Dave Stevenson (7):
      dt-bindings: display: bcm2711-hdmi: Add interrupt details for BCM2712
      dt-bindings: display: Fix BCM2835 HVS bindings for BCM2712
      dt-bindings: gpio: brcmstb: permit gpio-line-names property
      dt-bindings: interrupt-controller: brcm,bcm2836-l1-intc: Drop interrupt-controller requirement
      arm64: dts: broadcom: Rename bcm2712 interrupt controllers
      arm64: dts: broadcom: Correct hdmi device node names
      arm64: dts: broadcom: Remove intc controller on BCM2712.

 .../bindings/display/brcm,bcm2711-hdmi.yaml        | 107 ++++++++++++++++++---
 .../bindings/display/brcm,bcm2835-hvs.yaml         |  83 +++++++++++++---
 .../bindings/gpio/brcm,brcmstb-gpio.yaml           |   4 +
 .../interrupt-controller/brcm,bcm2836-l1-intc.yaml |   2 -
 arch/arm64/boot/dts/broadcom/bcm2712.dtsi          |  13 +--
 5 files changed, 170 insertions(+), 39 deletions(-)
---
base-commit: 3a6b7ba51f16c093420959ab2bd3476d180547fa
change-id: 20241128-dt-bcm2712-fixes-afb0e8a0a476

Best regards,
-- 
Dave Stevenson <dave.stevenson@xxxxxxxxxxxxxxx>