linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Wahren <stefan.wahren@i2se.com>
To: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
Cc: Dave Stevenson <dave.stevenson@raspberrypi.com>,
	devicetree@vger.kernel.org, kernel-list@raspberrypi.com,
	laurent.pinchart@ideasonboard.com,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, LMML <linux-media@vger.kernel.org>,
	linux-rpi-kernel@lists.infradead.org, lukasz@jany.st,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	naush@raspberrypi.com, Rob Herring <robh@kernel.org>,
	tomi.valkeinen@ideasonboard.com,
	Nicolas Saenz Julienne <nsaenz@kernel.org>,
	bcm-kernel-feedback-list@broadcom.com
Subject: Re: [RFC PATCH v4 05/12] media: bcm2835-unicam: Add support for CCP2/CSI2 camera interface
Date: Thu, 3 Feb 2022 21:42:21 +0100	[thread overview]
Message-ID: <eeac52c6-dc48-bf65-9844-a3d5ffb70215@i2se.com> (raw)
In-Reply-To: <20220203175009.558868-6-jeanmichel.hautbois@ideasonboard.com>

Hi Jean-Michel,

Am 03.02.22 um 18:50 schrieb Jean-Michel Hautbois:
> Add driver for the Unicam camera receiver block on BCM283x processors.
> It is represented as two video device nodes: unicam-image and
> unicam-embedded which are connected to an internal subdev (named
> unicam-subdev) in order to manage streams routing.
>
> Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
> Signed-off-by: Naushir Patuck <naush@raspberrypi.com>
> Signed-off-by: Jean-Michel Hautbois <jeanmichel.hautbois@ideasonboard.com>
>
> ---
> v3 main changes:
>   - Change code organization
>   - Remove unused variables
>   - Correct the fmt_meta functions
>   - Rewrite the start/stop streaming
>     - You can now start the image node alone, but not the metadata one
>     - The buffers are allocated per-node
>     - only the required stream is started, if the route exists and is
>       enabled
>   - Prefix the macros with UNICAM_ to not have too generic names
>   - Drop colorspace support
>     -> This is causing issues in the try-fmt v4l2-compliance test
>   test VIDIOC_G_FMT: OK
> 	fail: v4l2-test-formats.cpp(363): colorspace >= 0xff
> 	fail: v4l2-test-formats.cpp(465): testColorspace(!node->is_io_mc, pix.pixelformat, pix.colorspace, pix.ycbcr_enc, pix.quantization)
>   test VIDIOC_TRY_FMT: FAIL
> 	fail: v4l2-test-formats.cpp(363): colorspace >= 0xff
> 	fail: v4l2-test-formats.cpp(465): testColorspace(!node->is_io_mc, pix.pixelformat, pix.colorspace, pix.ycbcr_enc, pix.quantization)
>   test VIDIOC_S_FMT: FAIL
>
> v2: Remove the unicam_{info,debug,error} macros and use
> dev_dbg/dev_err instead.
> ---
>  MAINTAINERS                                   |    1 +
>  drivers/media/platform/Kconfig                |    1 +
>  drivers/media/platform/Makefile               |    2 +
>  drivers/media/platform/bcm2835/Kconfig        |   21 +
>  drivers/media/platform/bcm2835/Makefile       |    3 +
>  .../media/platform/bcm2835/bcm2835-unicam.c   | 2586 +++++++++++++++++
>  .../media/platform/bcm2835/vc4-regs-unicam.h  |  253 ++
>  7 files changed, 2867 insertions(+)
>  create mode 100644 drivers/media/platform/bcm2835/Kconfig
>  create mode 100644 drivers/media/platform/bcm2835/Makefile
>  create mode 100644 drivers/media/platform/bcm2835/bcm2835-unicam.c
>  create mode 100644 drivers/media/platform/bcm2835/vc4-regs-unicam.h
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 740cf86c56dc..ef2cdb37e06f 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -3676,6 +3676,7 @@ L:	linux-media@vger.kernel.org
>  S:	Maintained
>  F:	Documentation/devicetree/bindings/media/brcm,bcm2835-unicam.yaml
>  F:	arch/arm/boot/dts/bcm283x*
> +F:	drivers/media/platform/bcm2835/
>  
>  BROADCOM BCM47XX MIPS ARCHITECTURE
>  M:	Hauke Mehrtens <hauke@hauke-m.de>
Please move this change to the "single" MAINTAINER patch
> diff --git a/drivers/media/platform/Kconfig b/drivers/media/platform/Kconfig
> index 9fbdba0fd1e7..033b0358fbb8 100644
> --- a/drivers/media/platform/Kconfig
> +++ b/drivers/media/platform/Kconfig
> @@ -170,6 +170,7 @@ source "drivers/media/platform/am437x/Kconfig"
>  source "drivers/media/platform/xilinx/Kconfig"
>  source "drivers/media/platform/rcar-vin/Kconfig"
>  source "drivers/media/platform/atmel/Kconfig"
> +source "drivers/media/platform/bcm2835/Kconfig"
>  source "drivers/media/platform/sunxi/Kconfig"
>  
>  config VIDEO_TI_CAL
> diff --git a/drivers/media/platform/Makefile b/drivers/media/platform/Makefile
> index 28eb4aadbf45..18894fc586aa 100644
> --- a/drivers/media/platform/Makefile
> +++ b/drivers/media/platform/Makefile
> @@ -86,6 +86,8 @@ obj-$(CONFIG_VIDEO_QCOM_CAMSS)		+= qcom/camss/
>  
>  obj-$(CONFIG_VIDEO_QCOM_VENUS)		+= qcom/venus/
>  
> +obj-$(CONFIG_VIDEO_BCM2835_UNICAM)	+= bcm2835/
> +
>  obj-y					+= sunxi/
>  
>  obj-$(CONFIG_VIDEO_MESON_GE2D)		+= meson/ge2d/
> diff --git a/drivers/media/platform/bcm2835/Kconfig b/drivers/media/platform/bcm2835/Kconfig
> new file mode 100644
> index 000000000000..1691541da905
> --- /dev/null
> +++ b/drivers/media/platform/bcm2835/Kconfig
> @@ -0,0 +1,21 @@
> +# Broadcom VideoCore4 V4L2 camera support
> +
> +config VIDEO_BCM2835_UNICAM
> +	tristate "Broadcom BCM283x/BCM271x Unicam video capture driver"
> +	depends on ARCH_BCM2835 || COMPILE_TEST
> +	depends on VIDEO_V4L2
> +	select MEDIA_CONTROLLER
> +	select VIDEO_V4L2_SUBDEV_API
> +	select V4L2_FWNODE
> +	select VIDEOBUF2_DMA_CONTIG
> +	help
> +	  Say Y here to enable support for the BCM283x/BCM271x CSI-2 receiver.
> +	  This is a V4L2 driver that controls the CSI-2 receiver directly,
> +	  independently from the VC4 firmware.
> +	  This driver is mutually exclusive with the use of bcm2835-camera. The
> +	  firmware will disable all access to the peripheral from within the
> +	  firmware if it finds a DT node using it, and bcm2835-camera will
> +	  therefore fail to probe.
> +
> +	  To compile this driver as a module, choose M here. The module will be
> +	  called bcm2835-unicam.
> diff --git a/drivers/media/platform/bcm2835/Makefile b/drivers/media/platform/bcm2835/Makefile
> new file mode 100644
> index 000000000000..a98aba03598a
> --- /dev/null
> +++ b/drivers/media/platform/bcm2835/Makefile
> @@ -0,0 +1,3 @@
> +# Makefile for BCM2835 Unicam driver
> +
> +obj-$(CONFIG_VIDEO_BCM2835_UNICAM) += bcm2835-unicam.o
> diff --git a/drivers/media/platform/bcm2835/bcm2835-unicam.c b/drivers/media/platform/bcm2835/bcm2835-unicam.c
> new file mode 100644
> index 000000000000..95e12fdb90f5
> --- /dev/null
> +++ b/drivers/media/platform/bcm2835/bcm2835-unicam.c
> @@ -0,0 +1,2586 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * BCM283x / BCM271x Unicam Capture Driver
> + *
> + * Copyright (C) 2017-2020 - Raspberry Pi (Trading) Ltd.
> + *
> + * Dave Stevenson <dave.stevenson@raspberrypi.com>
> + *
> + * Based on TI am437x driver by
> + *   Benoit Parrot <bparrot@ti.com>
> + *   Lad, Prabhakar <prabhakar.csengg@gmail.com>
> + *
> + * and TI CAL camera interface driver by
> + *    Benoit Parrot <bparrot@ti.com>
> + *
> + *
> + * There are two camera drivers in the kernel for BCM283x - this one
> + * and bcm2835-camera (currently in staging).
> + *
> + * This driver directly controls the Unicam peripheral - there is no
> + * involvement with the VideoCore firmware. Unicam receives CSI-2 or
> + * CCP2 data and writes it into SDRAM.
> + * The only potential processing options are to repack Bayer data into an
> + * alternate format, and applying windowing.
> + * The repacking does not shift the data, so can repack V4L2_PIX_FMT_Sxxxx10P
> + * to V4L2_PIX_FMT_Sxxxx10, or V4L2_PIX_FMT_Sxxxx12P to V4L2_PIX_FMT_Sxxxx12,
> + * but not generically up to V4L2_PIX_FMT_Sxxxx16. The driver will add both
> + * formats where the relevant formats are defined, and will automatically
> + * configure the repacking as required.
> + * Support for windowing may be added later.
> + *
> + * It should be possible to connect this driver to any sensor with a
> + * suitable output interface and V4L2 subdevice driver.
> + *
> + * bcm2835-camera uses the VideoCore firmware to control the sensor,
> + * Unicam, ISP, and all tuner control loops. Fully processed frames are
> + * delivered to the driver by the firmware. It only has sensor drivers
> + * for Omnivision OV5647, and Sony IMX219 sensors.
> + *
> + * The two drivers are mutually exclusive for the same Unicam instance.
> + * The VideoCore firmware checks the device tree configuration during boot.
> + * If it finds device tree nodes called csi0 or csi1 it will block the
> + * firmware from accessing the peripheral, and bcm2835-camera will
> + * not be able to stream data.
> + */
> +
> +#include <linux/clk.h>
> +#include <linux/delay.h>
> +#include <linux/device.h>
> +#include <linux/dma-mapping.h>
> +#include <linux/err.h>
> +#include <linux/init.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/of_device.h>
> +#include <linux/of_graph.h>
> +#include <linux/pinctrl/consumer.h>
> +#include <linux/platform_device.h>
> +#include <linux/pm_runtime.h>
> +#include <linux/slab.h>
> +#include <linux/uaccess.h>
> +#include <linux/videodev2.h>
> +
> +#include <media/v4l2-async.h>
> +#include <media/v4l2-common.h>
> +#include <media/v4l2-ctrls.h>
> +#include <media/v4l2-dev.h>
> +#include <media/v4l2-device.h>
> +#include <media/v4l2-dv-timings.h>
> +#include <media/v4l2-event.h>
> +#include <media/v4l2-ioctl.h>
> +#include <media/v4l2-fwnode.h>
> +#include <media/v4l2-mc.h>
> +#include <media/videobuf2-dma-contig.h>
> +
> +#include "vc4-regs-unicam.h"
> +
> +#define UNICAM_MODULE_NAME	"unicam"
> +
> +/*
> + * Unicam must request a minimum of 250Mhz from the VPU clock.
> + * Otherwise the input FIFOs overrun and cause image corruption.
> + */
> +#define UNICAM_MIN_VPU_CLOCK_RATE (250 * 1000 * 1000)
> +
> +/*
> + * Stride is a 16 bit register, but also has to be a multiple of 32.
> + */
> +#define UNICAM_BPL_ALIGNMENT		32
> +#define UNICAM_MAX_BYTESPERLINE		((1 << 16) - UNICAM_BPL_ALIGNMENT)
> +/*
> + * Max width is therefore determined by the max stride divided by
> + * the number of bits per pixel. Take 32bpp as a
> + * worst case.
> + * No imposed limit on the height, so adopt a square image for want
> + * of anything better.
> + */
> +#define UNICAM_MAX_WIDTH		(UNICAM_MAX_BYTESPERLINE / 4)
> +#define UNICAM_MAX_HEIGHT		UNICAM_MAX_WIDTH
> +/* Define a nominal minimum image size */
> +#define UNICAM_MIN_WIDTH		16
> +#define UNICAM_MIN_HEIGHT		16
> +/* Default size of the embedded buffer */
> +#define UNICAM_EMBEDDED_SIZE	16384
> +
> +/*
> + * Size of the dummy buffer. Can be any size really, but the DMA
> + * allocation works in units of page sizes.
> + */
> +#define UNICAM_DUMMY_BUF_SIZE		PAGE_SIZE
> +
> +#define UNICAM_SD_PAD_SINK		0
> +#define UNICAM_SD_PAD_SOURCE_IMAGE	1
> +#define UNICAM_SD_PAD_SOURCE_METADATA	2
> +#define UNICAM_SD_NUM_PADS		(1 + UNICAM_SD_PAD_SOURCE_METADATA)
> +
> +static inline bool unicam_sd_pad_is_sink(u32 pad)
> +{
> +	/* Camera RX has 1 sink pad, and N source pads */
> +	return pad == 0;
> +}
> +
> +static inline bool unicam_sd_pad_is_source(u32 pad)
> +{
> +	/* Camera RX has 1 sink pad, and N source pads */
> +	return pad != UNICAM_SD_PAD_SINK;
> +}
> +
> +enum unicam_node_type {
> +	UNICAM_IMAGE_NODE,
> +	UNICAM_METADATA_NODE,
> +	UNICAM_MAX_NODES
> +};
> +
> +/* -----------------------------------------------------------------------------
> + * Structure definitions
> + */
Please remove those unnecessary comments ...
> +
> +/*
> + * struct unicam_fmt - Unicam media bus format information
> + * @fourcc: V4L2 pixel format FCC identifier. 0 if n/a.
> + * @unpacked_fourcc: V4L2 pixel format FCC identifier if the data is expanded
> + * out to 16bpp. 0 if n/a.
> + * @code: V4L2 media bus format code.
> + * @depth: Bits per pixel as delivered from the source.
> + * @csi_dt: CSI data type.
> + * @metadata_fmt: This format only applies to the metadata pad.
> + */
> +struct unicam_fmt {
> +	u32	fourcc;
> +	u32	unpacked_fourcc;
> +	u32	code;
> +	u8	depth;
> +	u8	csi_dt;
> +	u8	metadata_fmt:1;
> +};
> +
> +static const struct unicam_fmt formats[] = {
> +	/* YUV Formats */
> +	{
> +		.fourcc		= V4L2_PIX_FMT_YUYV,
> +		.code		= MEDIA_BUS_FMT_YUYV8_1X16,
> +		.depth		= 16,
> +		.csi_dt		= 0x1e,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_UYVY,
> +		.code		= MEDIA_BUS_FMT_UYVY8_1X16,
> +		.depth		= 16,
> +		.csi_dt		= 0x1e,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_YVYU,
> +		.code		= MEDIA_BUS_FMT_YVYU8_1X16,
> +		.depth		= 16,
> +		.csi_dt		= 0x1e,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_VYUY,
> +		.code		= MEDIA_BUS_FMT_VYUY8_1X16,
> +		.depth		= 16,
> +		.csi_dt		= 0x1e,
> +	}, {
> +	/* RGB Formats */
> +		.fourcc		= V4L2_PIX_FMT_RGB565, /* gggbbbbb rrrrrggg */
> +		.code		= MEDIA_BUS_FMT_RGB565_1X16,
> +		.depth		= 16,
> +		.csi_dt		= 0x22,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_RGB24, /* rgb */
> +		.code		= MEDIA_BUS_FMT_RGB888_1X24,
> +		.depth		= 24,
> +		.csi_dt		= 0x24,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_BGR24, /* bgr */
> +		.code		= MEDIA_BUS_FMT_BGR888_1X24,
> +		.depth		= 24,
> +		.csi_dt		= 0x24,
> +	}, {
> +	/* Bayer Formats */
> +		.fourcc		= V4L2_PIX_FMT_SBGGR8,
> +		.code		= MEDIA_BUS_FMT_SBGGR8_1X8,
> +		.depth		= 8,
> +		.csi_dt		= 0x2a,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGBRG8,
> +		.code		= MEDIA_BUS_FMT_SGBRG8_1X8,
> +		.depth		= 8,
> +		.csi_dt		= 0x2a,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGRBG8,
> +		.code		= MEDIA_BUS_FMT_SGRBG8_1X8,
> +		.depth		= 8,
> +		.csi_dt		= 0x2a,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SRGGB8,
> +		.code		= MEDIA_BUS_FMT_SRGGB8_1X8,
> +		.depth		= 8,
> +		.csi_dt		= 0x2a,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SBGGR10P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SBGGR10,
> +		.code		= MEDIA_BUS_FMT_SBGGR10_1X10,
> +		.depth		= 10,
> +		.csi_dt		= 0x2b,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGBRG10P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SGBRG10,
> +		.code		= MEDIA_BUS_FMT_SGBRG10_1X10,
> +		.depth		= 10,
> +		.csi_dt		= 0x2b,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGRBG10P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SGRBG10,
> +		.code		= MEDIA_BUS_FMT_SGRBG10_1X10,
> +		.depth		= 10,
> +		.csi_dt		= 0x2b,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SRGGB10P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SRGGB10,
> +		.code		= MEDIA_BUS_FMT_SRGGB10_1X10,
> +		.depth		= 10,
> +		.csi_dt		= 0x2b,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SBGGR12P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SBGGR12,
> +		.code		= MEDIA_BUS_FMT_SBGGR12_1X12,
> +		.depth		= 12,
> +		.csi_dt		= 0x2c,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGBRG12P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SGBRG12,
> +		.code		= MEDIA_BUS_FMT_SGBRG12_1X12,
> +		.depth		= 12,
> +		.csi_dt		= 0x2c,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGRBG12P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SGRBG12,
> +		.code		= MEDIA_BUS_FMT_SGRBG12_1X12,
> +		.depth		= 12,
> +		.csi_dt		= 0x2c,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SRGGB12P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SRGGB12,
> +		.code		= MEDIA_BUS_FMT_SRGGB12_1X12,
> +		.depth		= 12,
> +		.csi_dt		= 0x2c,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SBGGR14P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SBGGR14,
> +		.code		= MEDIA_BUS_FMT_SBGGR14_1X14,
> +		.depth		= 14,
> +		.csi_dt		= 0x2d,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGBRG14P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SGBRG14,
> +		.code		= MEDIA_BUS_FMT_SGBRG14_1X14,
> +		.depth		= 14,
> +		.csi_dt		= 0x2d,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SGRBG14P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SGRBG14,
> +		.code		= MEDIA_BUS_FMT_SGRBG14_1X14,
> +		.depth		= 14,
> +		.csi_dt		= 0x2d,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_SRGGB14P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_SRGGB14,
> +		.code		= MEDIA_BUS_FMT_SRGGB14_1X14,
> +		.depth		= 14,
> +		.csi_dt		= 0x2d,
> +	}, {
> +	/*
> +	 * 16 bit Bayer formats could be supported, but there is no CSI2
> +	 * data_type defined for raw 16, and no sensors that produce it at
> +	 * present.
> +	 */
> +
> +	/* Greyscale formats */
> +		.fourcc		= V4L2_PIX_FMT_GREY,
> +		.code		= MEDIA_BUS_FMT_Y8_1X8,
> +		.depth		= 8,
> +		.csi_dt		= 0x2a,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_Y10P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_Y10,
> +		.code		= MEDIA_BUS_FMT_Y10_1X10,
> +		.depth		= 10,
> +		.csi_dt		= 0x2b,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_Y12P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_Y12,
> +		.code		= MEDIA_BUS_FMT_Y12_1X12,
> +		.depth		= 12,
> +		.csi_dt		= 0x2c,
> +	}, {
> +		.fourcc		= V4L2_PIX_FMT_Y14P,
> +		.unpacked_fourcc = V4L2_PIX_FMT_Y14,
> +		.code		= MEDIA_BUS_FMT_Y14_1X14,
> +		.depth		= 14,
> +		.csi_dt		= 0x2d,
> +	},
> +	/* Embedded data format */
> +	{
> +		.fourcc		= V4L2_META_FMT_8,
> +		.code		= MEDIA_BUS_FMT_METADATA_8,
> +		.depth		= 8,
> +		.metadata_fmt	= 1,
> +	}
> +};
> +
> +struct unicam_buffer {
> +	struct vb2_v4l2_buffer vb;
> +	struct list_head list;
> +};
> +
> +static inline struct unicam_buffer *to_unicam_buffer(struct vb2_buffer *vb)
> +{
> +	return container_of(vb, struct unicam_buffer, vb.vb2_buf);
> +}
> +
> +struct unicam_node {
> +	bool registered;
> +	bool streaming;
> +	unsigned int node_id;
> +	/* Source pad id on the subdev for this node */
> +	unsigned int src_pad_id;
> +	/* Pointer pointing to current v4l2_buffer */
> +	struct unicam_buffer *cur_frm;
> +	/* Pointer pointing to next v4l2_buffer */
> +	struct unicam_buffer *next_frm;
> +	/* video capture */
> +	const struct unicam_fmt *fmt;
> +	/* Used to store current pixel format */
> +	struct v4l2_format v_fmt;
> +	/* Used to store current mbus frame format */
> +	struct v4l2_mbus_framefmt m_fmt;
> +	/* Buffer queue used in video-buf */
> +	struct vb2_queue buffer_queue;
> +	/* Queue of filled frames */
> +	struct list_head dma_queue;
> +	/* IRQ lock for DMA queue */
> +	spinlock_t dma_queue_lock;
> +	/* lock used to access this structure */
> +	struct mutex lock;
> +	/* Identifies video device for this channel */
> +	struct video_device video_dev;
> +	/* Pointer to the parent handle */
> +	struct unicam_device *dev;
> +	struct media_pad pad;
> +	struct media_pipeline pipe;
> +	/*
> +	 * Dummy buffer intended to be used by unicam
> +	 * if we have no other queued buffers to swap to.
> +	 */
> +	void *dummy_buf_cpu_addr;
> +	dma_addr_t dummy_buf_dma_addr;
> +
> +	/* Stream connected */
> +	u32 stream;
> +};
> +
> +struct unicam_device {
> +	struct kref kref;
> +
> +	/* V4l2 specific parameters */
> +	struct v4l2_async_subdev asd;
> +
> +	/* peripheral base address */
> +	void __iomem *base;
> +	/* clock gating base address */
> +	void __iomem *clk_gate_base;
> +	/* lp clock handle */
> +	struct clk *clock;
> +	/* vpu clock handle */
> +	struct clk *vpu_clock;
> +	/* V4l2 device */
> +	struct v4l2_device v4l2_dev;
> +	struct media_device mdev;
> +
> +	/* parent device */
> +	struct device *pdev;
> +	/* subdevice async Notifier */
> +	struct v4l2_async_notifier notifier;
> +	unsigned int sequence;
> +
> +	/* ptr to  sub device */
> +	struct v4l2_subdev *sensor;
> +
> +	/* Internal subdev */
> +	struct {
> +		struct v4l2_subdev sd;
> +		struct media_pad pads[UNICAM_SD_NUM_PADS];
> +		bool streaming;
> +	} subdev;
> +
> +	/* Remote pad pointer */
> +	struct media_pad *remote_pad;
> +
> +	enum v4l2_mbus_type bus_type;
> +	/*
> +	 * Stores bus.mipi_csi2.flags for CSI2 sensors, or
> +	 * bus.mipi_csi1.strobe for CCP2.
> +	 */
> +	unsigned int bus_flags;
> +	unsigned int max_data_lanes;
> +	unsigned int active_data_lanes;
> +	bool sensor_embedded_data;
> +
> +	struct unicam_node node[UNICAM_MAX_NODES];
> +};
> +
> +static inline struct unicam_device *
> +v4l2_device_to_unicam_device(struct v4l2_device *v4l2_dev)
> +{
> +	return container_of(v4l2_dev, struct unicam_device, v4l2_dev);
> +}
> +
> +static inline struct unicam_device *
> +sd_to_unicam_device(struct v4l2_subdev *sd)
> +{
> +	return container_of(sd, struct unicam_device, subdev.sd);
> +}
> +
> +/* -----------------------------------------------------------------------------
> + * Misc helper functions
> + */
> +
> +static inline bool is_metadata_node(struct unicam_node *node)
> +{
> +	return node->video_dev.device_caps & V4L2_CAP_META_CAPTURE;
> +}
> +
> +static inline bool is_image_node(struct unicam_node *node)
> +{
> +	return node->video_dev.device_caps & V4L2_CAP_VIDEO_CAPTURE;
> +}
> +
> +/* Hardware access */
> +static inline void unicam_clk_write(struct unicam_device *dev, u32 val)
> +{
> +	writel(val | 0x5a000000, dev->clk_gate_base);

This magic number has a special meaning for the clock manager. Look at
drivers/clk/bcm/clk-bcm2835.c. Either we use the define CM_PASSWORD or
add at least a comment.

...

> +
> +static int of_unicam_connect_subdevs(struct unicam_device *dev)
> +{
> +	struct device *pdev = dev->pdev;
> +	struct v4l2_fwnode_endpoint ep = { };
> +	struct fwnode_handle *ep_handle;
> +	struct v4l2_async_subdev *asd;
> +	unsigned int lane;
> +	int ret = -EINVAL;
> +
> +	if (of_property_read_u32(pdev->of_node, "num-data-lanes",
After adding the vendor prefix in the binding, i think this should be
updated?
> +				 &dev->max_data_lanes) < 0) {
> +		dev_err(dev->v4l2_dev.dev, "number of data lanes not set\n");
> +		return -EINVAL;
> +	}
> +
...
> +
> +static int unicam_probe(struct platform_device *pdev)
> +{
> +	struct unicam_device *unicam;
> +	int ret = 0;
> +
> +	unicam = kzalloc(sizeof(*unicam), GFP_KERNEL);
> +	if (!unicam)
> +		return -ENOMEM;
> +
> +	kref_init(&unicam->kref);
> +	unicam->pdev = &pdev->dev;
> +
> +	unicam->base = devm_platform_ioremap_resource(pdev, 0);

After adding the reg-names to the binding the driver can access the
register by name:

devm_platform_ioremap_resource_byname()

This is easier to read/understand.

Best regards



  reply	other threads:[~2022-02-03 20:44 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-03 17:49 [RFC PATCH v4 00/12] Add support for BCM2835 camera interface (unicam) Jean-Michel Hautbois
2022-02-03 17:49 ` [RFC PATCH v4 01/12] media: v4l: Add V4L2-PIX-FMT-Y12P format Jean-Michel Hautbois
2022-02-03 17:49 ` [RFC PATCH v4 02/12] media: v4l: Add V4L2-PIX-FMT-Y14P format Jean-Michel Hautbois
2022-02-03 17:50 ` [RFC PATCH v4 03/12] dt-bindings: media: Add bindings for bcm2835-unicam Jean-Michel Hautbois
2022-02-03 20:01   ` Stefan Wahren
2022-02-04  2:42   ` Laurent Pinchart
2022-02-04  8:50   ` (EXT) " Alexander Stein
2022-02-05  2:22     ` Laurent Pinchart
2022-02-07  6:30       ` (EXT) " Alexander Stein
2022-02-08  1:36         ` Laurent Pinchart
2022-02-08 11:35           ` Dave Stevenson
2022-02-03 17:50 ` [RFC PATCH v4 04/12] media: MAINTAINERS: add bcm2835 unicam driver Jean-Michel Hautbois
2022-02-04  2:40   ` Laurent Pinchart
2022-02-03 17:50 ` [RFC PATCH v4 05/12] media: bcm2835-unicam: Add support for CCP2/CSI2 camera interface Jean-Michel Hautbois
2022-02-03 20:42   ` Stefan Wahren [this message]
2022-02-05  2:20   ` Laurent Pinchart
2022-02-03 17:50 ` [RFC PATCH v4 06/12] ARM: dts: bcm2711: Add unicam CSI nodes Jean-Michel Hautbois
2022-02-03 20:06   ` Stefan Wahren
2022-02-04  2:45   ` Laurent Pinchart
2022-02-03 17:50 ` [RFC PATCH v4 07/12] media: imx219: Rename mbus codes array Jean-Michel Hautbois
2022-02-04  2:51   ` Laurent Pinchart
2022-02-03 17:50 ` [RFC PATCH v4 08/12] media: imx219: Switch from open to init_cfg Jean-Michel Hautbois
2022-02-04  2:54   ` Laurent Pinchart
2022-02-03 17:50 ` [RFC PATCH v4 09/12] media: imx219: Introduce the set_routing operation Jean-Michel Hautbois
2022-02-03 17:50 ` [RFC PATCH v4 10/12] media: imx219: use a local v4l2_subdev to simplify reading Jean-Michel Hautbois
2022-02-04  2:56   ` Laurent Pinchart
2022-02-03 17:50 ` [RFC PATCH v4 11/12] media: imx219: Add support for the V4L2 subdev active state Jean-Michel Hautbois
2022-02-03 17:50 ` [RFC PATCH v4 12/12] media: bcm283x: Include the imx219 node Jean-Michel Hautbois
2022-02-04  2:49   ` Laurent Pinchart
2022-02-03 19:55 ` [RFC PATCH v4 00/12] Add support for BCM2835 camera interface (unicam) Stefan Wahren
2022-02-04  0:38   ` Laurent Pinchart

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=eeac52c6-dc48-bf65-9844-a3d5ffb70215@i2se.com \
    --to=stefan.wahren@i2se.com \
    --cc=bcm-kernel-feedback-list@broadcom.com \
    --cc=dave.stevenson@raspberrypi.com \
    --cc=devicetree@vger.kernel.org \
    --cc=jeanmichel.hautbois@ideasonboard.com \
    --cc=kernel-list@raspberrypi.com \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-rpi-kernel@lists.infradead.org \
    --cc=lukasz@jany.st \
    --cc=mchehab@kernel.org \
    --cc=naush@raspberrypi.com \
    --cc=nsaenz@kernel.org \
    --cc=robh@kernel.org \
    --cc=tomi.valkeinen@ideasonboard.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).