drivers: media: pci: Update Hailo accelerator device driver to v4.20

Sourced from https://github.com/hailo-ai/hailort-drivers

Signed-off-by: Naushir Patuck <naush@raspberrypi.com>

overlays: Added waveshare 13.3inch panel support

Signed-off-by: eng33 <eng33@waveshare.com>

overlays: Added waveshare 13.3inch panel support

Signed-off-by: eng33 <eng33@waveshare.com>

Revert "DTS: bcm2712: enable SD slot CQE by default on Pi 5"

This reverts commit 216df57.

See #6349

There is an unknown hang when issuing flush commands that may be
triggered when IO is pending. Revert while investigation takes place.

configs: Really enable Hailo accelerator driver

For real, this time.

dts: bcm2712: cm5: Force the ANT pins to GPIOs

The magic required to make requesting a GPIO automatically change the
pinmux settings is currently absent. Work around that by explicitly
requesting that the ANT pins be in GPIO mode.

Signed-off-by: Phil Elwell <phil@raspberrypi.com>

drm/vc4: Fix atomic_async_check to call the right mode_set function

vc4_plane_atomic_async_check was always calling vc4_plane_mode_set
to validate and generate the dlist for the check. If async_check
decided it had to fall back to a sync commit, then this GEN4/5
dlist could get used on GEN6.

Call either vc4_plane_mode_set or vc6_plane_mode_set as appropriate.

Fixes: 1ab1fbb ("drm/vc4: hvs: Support BCM2712 HVS")
Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>

drm: vc4: Block swiotlb bounce buffers being imported as dmabuf

The dmabuf import already checks that the backing buffer is contiguous
and rejects it if it isn't. vc4 also requires that the buffer is
in the bottom 1GB of RAM, and this is all correctly defined via
dma-ranges.

However the kernel silently uses swiotlb to bounce dma buffers
around if they are in the wrong region. This relies on dma sync
functions to be called in order to copy the data to/from the
bounce buffer.

DRM is based on all memory allocations being coherent with the
GPU so that any updates to a framebuffer will be acted on without
the need for any additional update. This is fairly fundamentally
incompatible with needing to call dma_sync_ to handle the bounce
buffer copies, and therefore we have to detect and reject mappings
that use bounce buffers.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>

drivers: media: imx477: Set horizontal binning when disabling the scaler

The horizontal scaler has been disabled but actually the sensor is not
binning horizontally, resulting in images that are stretched 2x
horizontally (missing the right half of the field of view completely).

Therefore we must additionally set the horizontal binning mode. There
is only marginal change in output quality and noise levels.

Signed-off-by: David Plowman <david.plowman@raspberrypi.com>
Fixes: f075893 ("drivers: media: imx477: Disable the scaler")

fixup! overlays: mcp23017: allow specification of the i2c bus

The incorrect fragment order (*) caused broke the interrupt usage, and
while it was being fixed the lack of a reference to the pinctrl
declaration was noticed.

See: #5677

Signed-off-by: Phil Elwell <phil@raspberrypi.com>

(*) Ideally all fragments would appear in the file in the order in which
they should be merged, but that is easy to forget and can be awkward, so
the firmware merges all "intra" fragments (those that target other
fragments in the overlay) before "inter" fragments (those that target
the base DTB). However, intra fragments that target other intra
fragments is a level of nesting too far for this logic to cope, so they
must appear before the fragments they target.