| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <axda7onp2nm5225zxm2bpt3jrwoe4jdduhvtrckhdkjanjkmpu@uvezafny5ilg>
Date: Tue, 5 Nov 2024 11:06:30 +0100
From: Jacopo Mondi <jacopo.mondi@...asonboard.com>
To: Jacopo Mondi <jacopo.mondi@...asonboard.com>
Cc: keke.li@...ogic.com, Mauro Carvalho Chehab <mchehab@...nel.org>,
Rob Herring <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>,
Conor Dooley <conor+dt@...nel.org>, linux-media@...r.kernel.org, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org, kieran.bingham@...asonboard.com,
laurent.pinchart@...asonboard.com, dan.scally@...asonboard.com,
"tomi.valkeinen@...asonboard.com" <tomi.valkeinen@...asonboard.com>
Subject: Re: [PATCH v3 2/9] media: platform: Add c3 mipi csi2 driver
Hi again Keke
On Mon, Nov 04, 2024 at 06:50:11PM +0100, Jacopo Mondi wrote:
> Hi Keke
> sorry for the late feedback, hope you're still interested in
> upstreaming this driver
>
> On Wed, Sep 18, 2024 at 02:07:13PM +0800, Keke Li via B4 Relay wrote:
> > From: Keke Li <keke.li@...ogic.com>
> >
> > This driver is used to receive mipi data from image sensor.
> >
> > Reviewed-by: Daniel Scally <dan.scally@...asonboard.com>
> > Signed-off-by: Keke Li <keke.li@...ogic.com>
> > ---
> > MAINTAINERS | 7 +
> > drivers/media/platform/amlogic/Kconfig | 1 +
> > drivers/media/platform/amlogic/Makefile | 2 +
> > .../media/platform/amlogic/c3-mipi-csi2/Kconfig | 16 +
> > .../media/platform/amlogic/c3-mipi-csi2/Makefile | 3 +
> > .../platform/amlogic/c3-mipi-csi2/c3-mipi-csi2.c | 910 +++++++++++++++++++++
> > 6 files changed, 939 insertions(+)
> >
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index 2cdd7cacec86..9e75874a6e69 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -1209,6 +1209,13 @@ F: Documentation/devicetree/bindings/perf/amlogic,g12-ddr-pmu.yaml
> > F: drivers/perf/amlogic/
> > F: include/soc/amlogic/
> >
> > +AMLOGIC MIPI CSI2 DRIVER
> > +M: Keke Li <keke.li@...ogic.com>
> > +L: linux-media@...r.kernel.org
> > +S: Maintained
> > +F: Documentation/devicetree/bindings/media/amlogic,c3-mipi-csi2.yaml
> > +F: drivers/media/platform/amlogic/c3-mipi-csi2/
> > +
> > AMPHENOL CHIPCAP 2 HUMIDITY-TEMPERATURE IIO DRIVER
> > M: Javier Carrasco <javier.carrasco.cruz@...il.com>
> > L: linux-hwmon@...r.kernel.org
> > diff --git a/drivers/media/platform/amlogic/Kconfig b/drivers/media/platform/amlogic/Kconfig
> > index 5014957404e9..b7c2de14848b 100644
> > --- a/drivers/media/platform/amlogic/Kconfig
> > +++ b/drivers/media/platform/amlogic/Kconfig
> > @@ -2,4 +2,5 @@
> >
> > comment "Amlogic media platform drivers"
> >
> > +source "drivers/media/platform/amlogic/c3-mipi-csi2/Kconfig"
> > source "drivers/media/platform/amlogic/meson-ge2d/Kconfig"
> > diff --git a/drivers/media/platform/amlogic/Makefile b/drivers/media/platform/amlogic/Makefile
> > index d3cdb8fa4ddb..4f571ce5d13e 100644
> > --- a/drivers/media/platform/amlogic/Makefile
> > +++ b/drivers/media/platform/amlogic/Makefile
> > @@ -1,2 +1,4 @@
> > # SPDX-License-Identifier: GPL-2.0-only
> > +
> > +obj-y += c3-mipi-csi2/
> > obj-y += meson-ge2d/
> > diff --git a/drivers/media/platform/amlogic/c3-mipi-csi2/Kconfig b/drivers/media/platform/amlogic/c3-mipi-csi2/Kconfig
> > new file mode 100644
> > index 000000000000..0d7b2e203273
> > --- /dev/null
> > +++ b/drivers/media/platform/amlogic/c3-mipi-csi2/Kconfig
> > @@ -0,0 +1,16 @@
> > +# SPDX-License-Identifier: GPL-2.0-only
> > +
> > +config VIDEO_C3_MIPI_CSI2
> > + tristate "Amlogic C3 MIPI CSI-2 receiver"
> > + depends on ARCH_MESON || COMPILE_TEST
> > + depends on VIDEO_DEV
> > + depends on OF
> > + select MEDIA_CONTROLLER
> > + select V4L2_FWNODE
> > + select VIDEO_V4L2_SUBDEV_API
> > + help
> > + Video4Linux2 driver for Amlogic C3 MIPI CSI-2 receiver.
> > + C3 MIPI CSI-2 receiver is used to receive MIPI data from
> > + image sensor.
> > +
> > + To compile this driver as a module choose m here.
> > diff --git a/drivers/media/platform/amlogic/c3-mipi-csi2/Makefile b/drivers/media/platform/amlogic/c3-mipi-csi2/Makefile
> > new file mode 100644
> > index 000000000000..cc08fc722bfd
> > --- /dev/null
> > +++ b/drivers/media/platform/amlogic/c3-mipi-csi2/Makefile
> > @@ -0,0 +1,3 @@
> > +# SPDX-License-Identifier: GPL-2.0-only
> > +
> > +obj-$(CONFIG_VIDEO_C3_MIPI_CSI2) += c3-mipi-csi2.o
> > diff --git a/drivers/media/platform/amlogic/c3-mipi-csi2/c3-mipi-csi2.c b/drivers/media/platform/amlogic/c3-mipi-csi2/c3-mipi-csi2.c
> > new file mode 100644
> > index 000000000000..6ac60d5b26a8
> > --- /dev/null
> > +++ b/drivers/media/platform/amlogic/c3-mipi-csi2/c3-mipi-csi2.c
> > @@ -0,0 +1,910 @@
> > +// SPDX-License-Identifier: (GPL-2.0-only OR MIT)
> > +/*
> > + * Copyright (C) 2024 Amlogic, Inc. All rights reserved
> > + */
> > +
> > +#include <linux/cleanup.h>
> > +#include <linux/clk.h>
> > +#include <linux/device.h>
> > +#include <linux/module.h>
> > +#include <linux/mutex.h>
> > +#include <linux/platform_device.h>
> > +#include <linux/pm_runtime.h>
> > +
> > +#include <media/v4l2-async.h>
> > +#include <media/v4l2-common.h>
> > +#include <media/v4l2-device.h>
> > +#include <media/v4l2-fwnode.h>
> > +#include <media/v4l2-mc.h>
> > +#include <media/v4l2-subdev.h>
> > +
> > +/* C3 CSI-2 submodule definition */
> > +enum {
> > + SUBMD_APHY,
> > + SUBMD_DPHY,
> > + SUBMD_HOST,
> > +};
> > +
> > +#define CSI2_SUBMD_MASK GENMASK(17, 16)
> > +#define CSI2_SUBMD_SHIFT 16
> > +#define CSI2_SUBMD(x) (((x) & (CSI2_SUBMD_MASK)) >> (CSI2_SUBMD_SHIFT))
> > +#define CSI2_REG_ADDR_MASK GENMASK(15, 0)
> > +#define CSI2_REG_ADDR(x) ((x) & (CSI2_REG_ADDR_MASK))
> > +#define CSI2_REG_A(x) ((SUBMD_APHY << CSI2_SUBMD_SHIFT) | (x))
> > +#define CSI2_REG_D(x) ((SUBMD_DPHY << CSI2_SUBMD_SHIFT) | (x))
> > +#define CSI2_REG_H(x) ((SUBMD_HOST << CSI2_SUBMD_SHIFT) | (x))
> > +
> > +#define MIPI_CSI2_CLOCK_NUM_MAX 3
> > +#define MIPI_CSI2_SUBDEV_NAME "mipi-csi2"
>
> Isn't the name too generic ? Should it report at least "c3-mipi-csi2"
> ?
> > +
> > +/* C3 CSI-2 APHY register */
> > +#define MIPI_CSI_2M_PHY2_CNTL1 CSI2_REG_A(0x44)
> > +#define MIPI_APHY_NORMAL_CNTL1 0x3f425C00
>
> All other hex addresses use small capitals for letters
>
> > +
> > +#define MIPI_CSI_2M_PHY2_CNTL2 CSI2_REG_A(0x48)
> > +#define MIPI_APHY_4LANES_CNTL2 0x033a0000
> > +#define MIPI_APHY_NORMAL_CNTL2 0x333a0000
> > +
> > +#define MIPI_CSI_2M_PHY2_CNTL3 CSI2_REG_A(0x4c)
> > +#define MIPI_APHY_2LANES_CNTL3 0x03800000
> > +
> > +/* C3 CSI-2 DPHY register */
> > +#define MIPI_PHY_CTRL CSI2_REG_D(0x00)
> > +#define MIPI_DPHY_LANES_ENABLE 0x0
> > +
> > +#define MIPI_PHY_CLK_LANE_CTRL CSI2_REG_D(0x04)
> > +#define MIPI_DPHY_CLK_CONTINUE_MODE 0x3d8
> > +
> > +#define MIPI_PHY_DATA_LANE_CTRL CSI2_REG_D(0x08)
> > +#define MIPI_DPHY_LANE_CTRL_DISABLE 0x0
> > +
> > +#define MIPI_PHY_DATA_LANE_CTRL1 CSI2_REG_D(0x0c)
> > +#define MIPI_DPHY_INSERT_ERRESC BIT(0)
> > +#define MIPI_DPHY_HS_SYNC_CHECK BIT(1)
> > +#define MIPI_DPHY_FIVE_HS_PIPE GENMASK(6, 2)
> > +#define MIPI_DPHY_FIVE_HS_PIPE_SHIFT 2
> > +#define MIPI_DPHY_DATA_PIPE_SELECT GENMASK(9, 7)
> > +#define MIPI_DPHY_DATA_PIPE_SELECT_SHIFT 7
> > +
> > +#define MIPI_PHY_TCLK_MISS CSI2_REG_D(0x10)
> > +#define MIPI_DPHY_CLK_MISS 0x9
> > +
> > +#define MIPI_PHY_TCLK_SETTLE CSI2_REG_D(0x14)
> > +#define MIPI_DPHY_CLK_SETTLE 0x1F
> > +
> > +#define MIPI_PHY_THS_EXIT CSI2_REG_D(0x18)
> > +#define MIPI_DPHY_HS_EXIT 0x8
> > +
> > +#define MIPI_PHY_THS_SKIP CSI2_REG_D(0x1c)
> > +#define MIPI_DPHY_HS_SKIP 0xa
> > +
> > +#define MIPI_PHY_THS_SETTLE CSI2_REG_D(0x20)
> > +#define MIPI_PHY_TINIT CSI2_REG_D(0x24)
> > +#define MIPI_DPHY_INIT_CYCLES 0x4e20
> > +
> > +#define MIPI_PHY_TULPS_C CSI2_REG_D(0x28)
> > +#define MIPI_DPHY_ULPS_CHECK_CYCLES 0x1000
> > +
> > +#define MIPI_PHY_TULPS_S CSI2_REG_D(0x2c)
> > +#define MIPI_DPHY_ULPS_START_CYCLES 0x100
> > +
> > +#define MIPI_PHY_TMBIAS CSI2_REG_D(0x30)
> > +#define MIPI_DPHY_MBIAS_CYCLES 0x100
> > +
> > +#define MIPI_PHY_TLP_EN_W CSI2_REG_D(0x34)
> > +#define MIPI_DPHY_ULPS_STOP_CYCLES 0xC
> > +
> > +#define MIPI_PHY_TLPOK CSI2_REG_D(0x38)
> > +#define MIPI_DPHY_POWER_UP_CYCLES 0x100
> > +
> > +#define MIPI_PHY_TWD_INIT CSI2_REG_D(0x3c)
> > +#define MIPI_DPHY_INIT_WATCH_DOG 0x400000
> > +
> > +#define MIPI_PHY_TWD_HS CSI2_REG_D(0x40)
> > +#define MIPI_DPHY_HS_WATCH_DOG 0x400000
> > +
> > +#define MIPI_PHY_MUX_CTRL0 CSI2_REG_D(0x284)
> > +#define MIPI_DPHY_LANE3_SELECT GENMASK(3, 0)
> > +#define MIPI_DPHY_LANE2_SELECT GENMASK(7, 4)
> > +#define MIPI_DPHY_LANE2_SELECT_SHIFT 4
> > +#define MIPI_DPHY_LANE1_SELECT GENMASK(11, 8)
> > +#define MIPI_DPHY_LANE1_SELECT_SHIFT 8
> > +#define MIPI_DPHY_LANE0_SELECT GENMASK(14, 12)
> > +
> > +#define MIPI_PHY_MUX_CTRL1 CSI2_REG_D(0x288)
> > +#define MIPI_DPHY_LANE3_CTRL_SIGNAL GENMASK(3, 0)
> > +#define MIPI_DPHY_LANE2_CTRL_SIGNAL GENMASK(7, 4)
> > +#define MIPI_DPHY_LANE2_CTRL_SIGNAL_SHIFT 4
> > +#define MIPI_DPHY_LANE1_CTRL_SIGNAL GENMASK(11, 8)
> > +#define MIPI_DPHY_LANE1_CTRL_SIGNAL_SHIFT 8
> > +#define MIPI_DPHY_LANE0_CTRL_SIGNAL GENMASK(14, 12)
> > +#define MIPI_DPHY_CLK_SELECT BIT(17)
> > +
> > +/* C3 CSI-2 HOST register */
> > +#define CSI2_HOST_N_LANES CSI2_REG_H(0x04)
> > +#define CSI2_HOST_CSI2_RESETN CSI2_REG_H(0x10)
> > +#define CSI2_HOST_RESETN_DEFAULT 0x0
> > +#define CSI2_HOST_RESETN_RST_VALUE BIT(0)
> > +
> > +#define CSI2_HOST_MASK1 CSI2_REG_H(0x28)
> > +#define CSI2_HOST_ERROR_MASK1 GENMASK(28, 0)
> > +
> > +#define MIPI_CSI2_MAX_WIDTH 2888
> > +#define MIPI_CSI2_MIN_WIDTH 160
> > +#define MIPI_CSI2_MAX_HEIGHT 2240
> > +#define MIPI_CSI2_MIN_HEIGHT 120
> > +#define MIPI_CSI2_DEFAULT_WIDTH 1920
> > +#define MIPI_CSI2_DEFAULT_HEIGHT 1080
> > +#define MIPI_CSI2_DEFAULT_FMT MEDIA_BUS_FMT_SRGGB10_1X10
> > +
> > +/* C3 CSI-2 pad list */
> > +enum {
> > + MIPI_CSI2_PAD_SINK,
> > + MIPI_CSI2_PAD_SRC,
> > + MIPI_CSI2_PAD_MAX
> > +};
> > +
> > +/**
>
> You don't need to kernel-doc in-driver types and functions.
> Documentation is always good, but this won't be parsed by kernel-doc
> (afaiu) so you should drop one * from /**
>
> > + * struct csi_info - MIPI CSI2 information
> > + *
> > + * @clocks: array of MIPI CSI2 clock names
> > + * @clock_rates: array of MIPI CSI2 clock rate
> > + * @clock_num: actual clock number
> > + */
> > +struct csi_info {
> > + char *clocks[MIPI_CSI2_CLOCK_NUM_MAX];
> > + u32 clock_rates[MIPI_CSI2_CLOCK_NUM_MAX];
> > + u32 clock_num;
> > +};
> > +
> > +/**
> > + * struct csi_device - MIPI CSI2 platform device
> > + *
> > + * @dev: pointer to the struct device
> > + * @aphy: MIPI CSI2 aphy register address
> > + * @dphy: MIPI CSI2 dphy register address
> > + * @host: MIPI CSI2 host register address
> > + * @clks: array of MIPI CSI2 clocks
> > + * @sd: MIPI CSI2 sub-device
> > + * @pads: MIPI CSI2 sub-device pads
> > + * @notifier: notifier to register on the v4l2-async API
> > + * @src_sd: source sub-device
> > + * @bus: MIPI CSI2 bus information
> > + * @src_sd_pad: source sub-device pad
> > + * @lock: protect MIPI CSI2 device
> > + * @info: version-specific MIPI CSI2 information
> > + */
> > +struct csi_device {
> > + struct device *dev;
> > + void __iomem *aphy;
> > + void __iomem *dphy;
> > + void __iomem *host;
> > + struct clk_bulk_data clks[MIPI_CSI2_CLOCK_NUM_MAX];
> > +
> > + struct v4l2_subdev sd;
> > + struct media_pad pads[MIPI_CSI2_PAD_MAX];
> > + struct v4l2_async_notifier notifier;
> > + struct v4l2_subdev *src_sd;
> > + struct v4l2_mbus_config_mipi_csi2 bus;
> > +
> > + u16 src_sd_pad;
> > + struct mutex lock; /* Protect csi device */
>
> All the operations which receive a subdev_state are guaranteed to be locked
> so you can avoid manually locking in enable/disable streams
> (and drop #include cleanup.h if you don't use guards in any other
> place)
>
> > + const struct csi_info *info;
> > +};
> > +
> > +static const u32 c3_mipi_csi_formats[] = {
> > + MEDIA_BUS_FMT_SBGGR10_1X10,
> > + MEDIA_BUS_FMT_SGBRG10_1X10,
> > + MEDIA_BUS_FMT_SGRBG10_1X10,
> > + MEDIA_BUS_FMT_SRGGB10_1X10,
> > + MEDIA_BUS_FMT_SBGGR12_1X12,
> > + MEDIA_BUS_FMT_SGBRG12_1X12,
> > + MEDIA_BUS_FMT_SGRBG12_1X12,
> > + MEDIA_BUS_FMT_SRGGB12_1X12,
> > +};
> > +
> > +/* Hardware configuration */
> > +
> > +static void c3_mipi_csi_write(struct csi_device *csi, u32 reg, u32 val)
> > +{
> > + void __iomem *addr;
> > +
> > + switch (CSI2_SUBMD(reg)) {
> > + case SUBMD_APHY:
> > + addr = csi->aphy + CSI2_REG_ADDR(reg);
> > + break;
> > + case SUBMD_DPHY:
> > + addr = csi->dphy + CSI2_REG_ADDR(reg);
> > + break;
> > + case SUBMD_HOST:
> > + addr = csi->host + CSI2_REG_ADDR(reg);
> > + break;
> > + default:
> > + dev_err(csi->dev, "Invalid sub-module: %lu\n", CSI2_SUBMD(reg));
> > + return;
> > + }
> > +
> > + writel(val, addr);
> > +}
> > +
> > +static void c3_mipi_csi_update_bits(struct csi_device *csi, u32 reg,
> > + u32 mask, u32 val)
> > +{
> > + void __iomem *addr;
> > + u32 orig, tmp;
> > +
> > + switch (CSI2_SUBMD(reg)) {
> > + case SUBMD_APHY:
> > + addr = csi->aphy + CSI2_REG_ADDR(reg);
> > + break;
> > + case SUBMD_DPHY:
> > + addr = csi->dphy + CSI2_REG_ADDR(reg);
> > + break;
> > + case SUBMD_HOST:
> > + addr = csi->host + CSI2_REG_ADDR(reg);
> > + break;
> > + default:
> > + dev_err(csi->dev, "Invalid sub-module: %lu\n", CSI2_SUBMD(reg));
> > + return;
> > + }
>
> This is repeated in two functions and could be grouped to a common
> place. Up to you
>
> > +
> > + orig = readl(addr);
> > + tmp = orig & ~mask;
> > + tmp |= val & mask;
> > +
> > + if (tmp != orig)
> > + writel(tmp, addr);
> > +}
> > +
> > +static void c3_mipi_csi_cfg_aphy(struct csi_device *csi, u32 lanes)
> > +{
> > + c3_mipi_csi_write(csi, MIPI_CSI_2M_PHY2_CNTL1, MIPI_APHY_NORMAL_CNTL1);
> > +
> > + if (lanes == 4)
> > + c3_mipi_csi_write(csi, MIPI_CSI_2M_PHY2_CNTL2, MIPI_APHY_4LANES_CNTL2);
> > + else
> > + c3_mipi_csi_write(csi, MIPI_CSI_2M_PHY2_CNTL2, MIPI_APHY_NORMAL_CNTL2);
> > +
> > + if (lanes == 2)
> > + c3_mipi_csi_write(csi, MIPI_CSI_2M_PHY2_CNTL3, MIPI_APHY_2LANES_CNTL3);
>
> The driver seems to only accept 2 or 4 lanes. What is
> MIPI_APHY_NORMAL_CNTL2 for ?
>
> > +}
> > +
> > +static void c3_mipi_csi_2lanes_setting(struct csi_device *csi)
> > +{
> > + /* Disable lane 2 and lane 3 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE3_SELECT, 0xf);
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE2_SELECT,
> > + 0xf << MIPI_DPHY_LANE2_SELECT_SHIFT);
> > + /* Select analog data lane 1 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE1_SELECT,
> > + 0x1 << MIPI_DPHY_LANE1_SELECT_SHIFT);
> > + /* Select analog data lane 0 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE0_SELECT, 0x0);
> > +
> > + /* Disable lane 2 and lane 3 control signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE3_CTRL_SIGNAL, 0xf);
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE2_CTRL_SIGNAL,
> > + 0xf << MIPI_DPHY_LANE2_CTRL_SIGNAL_SHIFT);
> > + /* Select lane 1 signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE1_CTRL_SIGNAL,
> > + 0x1 << MIPI_DPHY_LANE1_CTRL_SIGNAL_SHIFT);
> > + /* Select lane 0 signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE0_CTRL_SIGNAL, 0x0);
> > + /* Select input 0 as clock */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_CLK_SELECT,
> > + MIPI_DPHY_CLK_SELECT);
> > +}
> > +
> > +static void c3_mipi_csi_4lanes_setting(struct csi_device *csi)
> > +{
> > + /* Select analog data lane 3 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE3_SELECT, 0x3);
> > + /* Select analog data lane 2 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE2_SELECT,
> > + 0x2 << MIPI_DPHY_LANE2_SELECT_SHIFT);
> > + /* Select analog data lane 1 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE1_SELECT,
> > + 0x1 << MIPI_DPHY_LANE1_SELECT_SHIFT);
> > + /* Select analog data lane 0 */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL0, MIPI_DPHY_LANE0_SELECT, 0x0);
> > +
> > + /* Select lane 3 signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE3_CTRL_SIGNAL, 0x3);
> > + /* Select lane 2 signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE2_CTRL_SIGNAL,
> > + 0x2 << MIPI_DPHY_LANE2_CTRL_SIGNAL_SHIFT);
> > + /* Select lane 1 signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE1_CTRL_SIGNAL,
> > + 0x1 << MIPI_DPHY_LANE1_CTRL_SIGNAL_SHIFT);
> > + /* Select lane 0 signal */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_LANE0_CTRL_SIGNAL, 0x0);
> > + /* Select input 0 as clock */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_MUX_CTRL1, MIPI_DPHY_CLK_SELECT,
> > + MIPI_DPHY_CLK_SELECT);
> > +}
> > +
> > +static void c3_mipi_csi_cfg_dphy(struct csi_device *csi, u32 lanes, s64 rate)
> > +{
> > + u32 val;
> > + u32 settle;
> > +
> > + /* Calculate the high speed settle */
> > + val = DIV_ROUND_UP(1000000000, rate);
> > + settle = (16 * val + 230) / 10;
> > +
> > + c3_mipi_csi_write(csi, MIPI_PHY_CLK_LANE_CTRL, MIPI_DPHY_CLK_CONTINUE_MODE);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TCLK_MISS, MIPI_DPHY_CLK_MISS);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TCLK_SETTLE, MIPI_DPHY_CLK_SETTLE);
> > + c3_mipi_csi_write(csi, MIPI_PHY_THS_EXIT, MIPI_DPHY_HS_EXIT);
> > + c3_mipi_csi_write(csi, MIPI_PHY_THS_SKIP, MIPI_DPHY_HS_SKIP);
> > + c3_mipi_csi_write(csi, MIPI_PHY_THS_SETTLE, settle);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TINIT, MIPI_DPHY_INIT_CYCLES);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TMBIAS, MIPI_DPHY_MBIAS_CYCLES);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TULPS_C, MIPI_DPHY_ULPS_CHECK_CYCLES);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TULPS_S, MIPI_DPHY_ULPS_START_CYCLES);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TLP_EN_W, MIPI_DPHY_ULPS_STOP_CYCLES);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TLPOK, MIPI_DPHY_POWER_UP_CYCLES);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TWD_INIT, MIPI_DPHY_INIT_WATCH_DOG);
> > + c3_mipi_csi_write(csi, MIPI_PHY_TWD_HS, MIPI_DPHY_HS_WATCH_DOG);
> > + c3_mipi_csi_write(csi, MIPI_PHY_DATA_LANE_CTRL, MIPI_DPHY_LANE_CTRL_DISABLE);
> > +
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_DATA_LANE_CTRL1, MIPI_DPHY_INSERT_ERRESC,
> > + MIPI_DPHY_INSERT_ERRESC);
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_DATA_LANE_CTRL1, MIPI_DPHY_HS_SYNC_CHECK,
> > + MIPI_DPHY_HS_SYNC_CHECK);
> > + /*
> > + * Set 5 pipe lines to the same high speed.
> > + * Each bit for one pipe line.
> > + */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_DATA_LANE_CTRL1, MIPI_DPHY_FIVE_HS_PIPE,
> > + 0x1f << MIPI_DPHY_FIVE_HS_PIPE_SHIFT);
> > +
> > + /* Output data with pipe line data. */
> > + c3_mipi_csi_update_bits(csi, MIPI_PHY_DATA_LANE_CTRL1, MIPI_DPHY_DATA_PIPE_SELECT,
> > + 0x3 << MIPI_DPHY_DATA_PIPE_SELECT_SHIFT);
>
> Would it be possible to provide a definition for these 0x1f and 0x3
> values ?
>
> > + if (lanes == 2)
> > + c3_mipi_csi_2lanes_setting(csi);
> > + else
> > + c3_mipi_csi_4lanes_setting(csi);
> > +
> > + /* Enable digital data and clock lanes */
> > + c3_mipi_csi_write(csi, MIPI_PHY_CTRL, MIPI_DPHY_LANES_ENABLE);
> > +}
> > +
> > +static void c3_mipi_csi_cfg_host(struct csi_device *csi, u32 lanes)
> > +{
> > + /* Reset CSI-2 controller output */
> > + c3_mipi_csi_write(csi, CSI2_HOST_CSI2_RESETN, CSI2_HOST_RESETN_DEFAULT);
> > + c3_mipi_csi_write(csi, CSI2_HOST_CSI2_RESETN, CSI2_HOST_RESETN_RST_VALUE);
> > +
> > + /* Set data lane number */
> > + c3_mipi_csi_write(csi, CSI2_HOST_N_LANES, lanes - 1);
> > +
> > + /* Enable error mask */
> > + c3_mipi_csi_write(csi, CSI2_HOST_MASK1, CSI2_HOST_ERROR_MASK1);
> > +}
> > +
> > +static int c3_mipi_csi_start_stream(struct csi_device *csi)
> > +{
> > + s64 link_freq;
> > + s64 lane_rate;
> > +
> > + link_freq = v4l2_get_link_freq(csi->src_sd->ctrl_handler, 0, 0);
> > + if (link_freq < 0) {
> > + dev_err(csi->dev, "Unable to obtain link frequency: %lld\n", link_freq);
> > + return link_freq;
> > + }
> > +
> > + lane_rate = link_freq * 2;
> > + if (lane_rate > 1500000000)
>
> I would dev_err here too
>
> > + return -EINVAL;
> > +
> > + c3_mipi_csi_cfg_aphy(csi, csi->bus.num_data_lanes);
> > + c3_mipi_csi_cfg_dphy(csi, csi->bus.num_data_lanes, lane_rate);
> > + c3_mipi_csi_cfg_host(csi, csi->bus.num_data_lanes);
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_enable_streams(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state,
> > + u32 pad, u64 streams_mask)
> > +{
> > + struct csi_device *csi = v4l2_get_subdevdata(sd);
> > + u64 sink_streams;
> > + int ret;
> > +
> > + guard(mutex)(&csi->lock);
> > +
> > + pm_runtime_resume_and_get(csi->dev);
> > +
> > + c3_mipi_csi_start_stream(csi);
> > +
> > + sink_streams = v4l2_subdev_state_xlate_streams(state, pad,
> > + MIPI_CSI2_PAD_SINK,
> > + &streams_mask);
> > + ret = v4l2_subdev_enable_streams(csi->src_sd,
> > + csi->src_sd_pad,
> > + sink_streams);
> > + if (ret) {
> > + pm_runtime_put(csi->dev);
> > + return ret;
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_disable_streams(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state,
> > + u32 pad, u64 streams_mask)
> > +{
> > + struct csi_device *csi = v4l2_get_subdevdata(sd);
> > + u64 sink_streams;
> > + int ret;
> > +
> > + guard(mutex)(&csi->lock);
> > +
> > + sink_streams = v4l2_subdev_state_xlate_streams(state, pad,
> > + MIPI_CSI2_PAD_SINK,
> > + &streams_mask);
> > + ret = v4l2_subdev_disable_streams(csi->src_sd,
> > + csi->src_sd_pad,
> > + sink_streams);
> > + if (ret)
> > + dev_err(csi->dev, "Failed to disable %s\n", csi->src_sd->name);
> > +
> > + pm_runtime_put(csi->dev);
> > +
> > + return ret;
> > +}
> > +
> > +static int c3_mipi_csi_cfg_routing(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state,
> > + struct v4l2_subdev_krouting *routing)
> > +{
> > + static const struct v4l2_mbus_framefmt format = {
> > + .width = MIPI_CSI2_DEFAULT_WIDTH,
> > + .height = MIPI_CSI2_DEFAULT_HEIGHT,
> > + .code = MIPI_CSI2_DEFAULT_FMT,
> > + .field = V4L2_FIELD_NONE,
> > + .colorspace = V4L2_COLORSPACE_RAW,
> > + .ycbcr_enc = V4L2_YCBCR_ENC_601,
> > + .quantization = V4L2_QUANTIZATION_LIM_RANGE,
>
> I presume for Raw Bayer data the quantization range is full ?
>
> > + .xfer_func = V4L2_XFER_FUNC_NONE,
> > + };
> > + int ret;
> > +
> > + ret = v4l2_subdev_routing_validate(sd, routing,
> > + V4L2_SUBDEV_ROUTING_ONLY_1_TO_1);
> > + if (ret)
> > + return ret;
>
> You should validate that the provided routing table matches what the
> driver supports, so only [0/0]->[1/0]
>
> Now that I've said so, if the routing table is not modifiable I wonder
> if you should support set_routing() at all, or it could be left out
> until you don't add support for more streams to the driver.
>
> After all this driver implements support for routing but doesn't set
> the V4L2_SUBDEV_FL_STREAMS flag, so the operation is disallowed from
> userspace for now.
>
> > +
> > + ret = v4l2_subdev_set_routing_with_fmt(sd, state, routing, &format);
> > + if (ret)
> > + return ret;
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_init_routing(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state)
> > +{
> > + struct v4l2_subdev_route routes;
> > + struct v4l2_subdev_krouting routing;
> > +
> > + routes.sink_pad = MIPI_CSI2_PAD_SINK;
> > + routes.sink_stream = 0;
> > + routes.source_pad = MIPI_CSI2_PAD_SRC;
> > + routes.source_stream = 0;
> > + routes.flags = V4L2_SUBDEV_ROUTE_FL_ACTIVE;
> > +
> > + routing.num_routes = 1;
> > + routing.routes = &routes;
> > +
> > + return c3_mipi_csi_cfg_routing(sd, state, &routing);
> > +}
> > +
> > +static int c3_mipi_csi_set_routing(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state,
> > + enum v4l2_subdev_format_whence which,
> > + struct v4l2_subdev_krouting *routing)
> > +{
> > + bool is_streaming = v4l2_subdev_is_streaming(sd);
> > +
> > + if (which == V4L2_SUBDEV_FORMAT_ACTIVE && is_streaming)
> > + return -EBUSY;
> > +
> > + return c3_mipi_csi_cfg_routing(sd, state, routing);
> > +}
> > +
> > +static int c3_mipi_csi_enum_mbus_code(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state,
> > + struct v4l2_subdev_mbus_code_enum *code)
> > +{
> > + switch (code->pad) {
> > + case MIPI_CSI2_PAD_SINK:
> > + if (code->index >= ARRAY_SIZE(c3_mipi_csi_formats))
> > + return -EINVAL;
> > +
> > + code->code = c3_mipi_csi_formats[code->index];
> > + break;
> > + case MIPI_CSI2_PAD_SRC:
> > + struct v4l2_mbus_framefmt *fmt;
> > +
> > + if (code->index > 0)
> > + return -EINVAL;
> > +
> > + fmt = v4l2_subdev_state_get_format(state, code->pad);
> > + code->code = fmt->code;
> > + break;
>
> I'm not sure if the V4L2 API specify that the formats on a pad should
> be enumerated in full, regardless of the configuration, or like you're
> doing here reflect the subdev configuration. I like what you have here
> more, so unless someone screams I think it's fine.
>
> > + default:
> > + return -EINVAL;
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_set_fmt(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state,
> > + struct v4l2_subdev_format *format)
> > +{
> > + struct v4l2_mbus_framefmt *fmt;
> > + unsigned int i;
> > +
> > + if (format->pad != MIPI_CSI2_PAD_SINK)
> > + return v4l2_subdev_get_fmt(sd, state, format);
> > +
> > + fmt = v4l2_subdev_state_get_format(state, format->pad);
>
> Could you clarify what other streams you plan to support ? As you
> support routing I presume you are preparing to capture
> multiple streams of data like image + embedded data, or to support
> sensors which sends data on different virtual channels ?
>
> How do you see this driver evolve ? Will it be augmented with an
> additional source pad directed to a video device where to capture
> embedded data from ?
>
> I'm wondering because if PAD_SINK become multiplexed, you won't be
> allowed to set a format there. It only works now because you have a
> single stream.
Pardon me, I've been made to notice the stream API allows userspace to
set a format on a [pad/stream] combination, so the above statement is
not correct: you will still be allowed to set a format a multiplexed
sink pad.
Knowing however how you plan to expand this to support multiple
streams is still something I would be interested in :)
Thanks
j
>
> Speaking of which, as you prepare to support multiple streams, I would
> specify the stream number when calling v4l2_subdev_state_get_format().
>
> fmt = v4l2_subdev_state_get_format(state, format->pad, 0);
>
> > +
> > + for (i = 0; i < ARRAY_SIZE(c3_mipi_csi_formats); i++)
> > + if (format->format.code == c3_mipi_csi_formats[i])
> > + break;
>
> nit: please use {} for the for loop
>
> > +
> > + if (i == ARRAY_SIZE(c3_mipi_csi_formats))
> > + fmt->code = c3_mipi_csi_formats[0];
> > + else
> > + fmt->code = c3_mipi_csi_formats[i];
>
> You could set this in the for loop, before breaking.
>
> > +
> > + fmt->width = clamp_t(u32, format->format.width,
> > + MIPI_CSI2_MIN_WIDTH, MIPI_CSI2_MAX_WIDTH);
> > + fmt->height = clamp_t(u32, format->format.height,
> > + MIPI_CSI2_MIN_HEIGHT, MIPI_CSI2_MAX_HEIGHT);
> > +
> You should set the colorspace related information too, as you
> initialize them, similar to what you do in init_state()
>
> > + format->format = *fmt;
> > +
> > + /* Synchronize the format to source pad */
> > + fmt = v4l2_subdev_state_get_format(state, MIPI_CSI2_PAD_SRC);
> > + *fmt = format->format;
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_init_state(struct v4l2_subdev *sd,
> > + struct v4l2_subdev_state *state)
> > +{
> > + struct v4l2_mbus_framefmt *sink_fmt;
> > + struct v4l2_mbus_framefmt *src_fmt;
> > +
> > + sink_fmt = v4l2_subdev_state_get_format(state, MIPI_CSI2_PAD_SINK);
> > + src_fmt = v4l2_subdev_state_get_format(state, MIPI_CSI2_PAD_SRC);
> > +
> > + sink_fmt->width = MIPI_CSI2_DEFAULT_WIDTH;
> > + sink_fmt->height = MIPI_CSI2_DEFAULT_HEIGHT;
> > + sink_fmt->field = V4L2_FIELD_NONE;
> > + sink_fmt->code = MIPI_CSI2_DEFAULT_FMT;
> > + sink_fmt->colorspace = V4L2_COLORSPACE_RAW;
> > + sink_fmt->xfer_func = V4L2_MAP_XFER_FUNC_DEFAULT(sink_fmt->colorspace);
> > + sink_fmt->ycbcr_enc = V4L2_MAP_YCBCR_ENC_DEFAULT(sink_fmt->colorspace);
> > + sink_fmt->quantization =
> > + V4L2_MAP_QUANTIZATION_DEFAULT(false, sink_fmt->colorspace,
>
> If you could initialize them like you do above, with specific values
> instead of using _DEFAULT() I think it's better.
>
> > + sink_fmt->ycbcr_enc);
> > + *src_fmt = *sink_fmt;
> > +
> > + return c3_mipi_csi_init_routing(sd, state);
> > +}
> > +
> > +static const struct v4l2_subdev_pad_ops c3_mipi_csi_pad_ops = {
> > + .enum_mbus_code = c3_mipi_csi_enum_mbus_code,
> > + .get_fmt = v4l2_subdev_get_fmt,
> > + .set_fmt = c3_mipi_csi_set_fmt,
> > + .set_routing = c3_mipi_csi_set_routing,
> > + .enable_streams = c3_mipi_csi_enable_streams,
> > + .disable_streams = c3_mipi_csi_disable_streams,
> > +};
> > +
> > +static const struct v4l2_subdev_ops c3_mipi_csi_subdev_ops = {
> > + .pad = &c3_mipi_csi_pad_ops,
> > +};
> > +
> > +static const struct v4l2_subdev_internal_ops c3_mipi_csi_internal_ops = {
> > + .init_state = c3_mipi_csi_init_state,
> > +};
> > +
> > +/* Media entity operations */
> > +static const struct media_entity_operations c3_mipi_csi_entity_ops = {
> > + .link_validate = v4l2_subdev_link_validate,
> > +};
> > +
> > +/* PM runtime */
> > +
> > +static int __maybe_unused c3_mipi_csi_runtime_suspend(struct device *dev)
> > +{
> > + struct csi_device *csi = dev_get_drvdata(dev);
> > +
> > + clk_bulk_disable_unprepare(csi->info->clock_num, csi->clks);
> > +
> > + return 0;
> > +}
> > +
> > +static int __maybe_unused c3_mipi_csi_runtime_resume(struct device *dev)
> > +{
> > + struct csi_device *csi = dev_get_drvdata(dev);
> > +
> > + return clk_bulk_prepare_enable(csi->info->clock_num, csi->clks);
> > +}
> > +
> > +static const struct dev_pm_ops c3_mipi_csi_pm_ops = {
> > + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend,
> > + pm_runtime_force_resume)
> > + SET_RUNTIME_PM_OPS(c3_mipi_csi_runtime_suspend,
> > + c3_mipi_csi_runtime_resume, NULL)
>
> You could use SYSTEM_SLEEP_PM_OPS and RUNTIME_PM_OPS and set
>
> .pm = pm_ptr(&c3_mipi_csi_pm_ops),
>
> to avoid __maybe_unused in the functions, up to you
>
> > +};
> > +
> > +/* Probe/remove & platform driver */
> > +
> > +static int c3_mipi_csi_subdev_init(struct csi_device *csi)
> > +{
> > + struct v4l2_subdev *sd = &csi->sd;
> > + int ret;
> > +
> > + v4l2_subdev_init(sd, &c3_mipi_csi_subdev_ops);
> > + sd->owner = THIS_MODULE;
> > + sd->flags |= V4L2_SUBDEV_FL_HAS_DEVNODE;
> > + sd->internal_ops = &c3_mipi_csi_internal_ops;
> > + snprintf(sd->name, sizeof(sd->name), "%s", MIPI_CSI2_SUBDEV_NAME);
> > +
> > + sd->entity.function = MEDIA_ENT_F_VID_IF_BRIDGE;
> > + sd->entity.ops = &c3_mipi_csi_entity_ops;
> > +
> > + sd->dev = csi->dev;
> > + v4l2_set_subdevdata(sd, csi);
> > +
> > + csi->pads[MIPI_CSI2_PAD_SINK].flags = MEDIA_PAD_FL_SINK;
> > + csi->pads[MIPI_CSI2_PAD_SRC].flags = MEDIA_PAD_FL_SOURCE;
> > + ret = media_entity_pads_init(&sd->entity, MIPI_CSI2_PAD_MAX, csi->pads);
> > + if (ret)
> > + return ret;
> > +
> > + ret = v4l2_subdev_init_finalize(sd);
> > + if (ret) {
> > + media_entity_cleanup(&sd->entity);
> > + return ret;
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +static void c3_mipi_csi_subdev_deinit(struct csi_device *csi)
> > +{
> > + v4l2_subdev_cleanup(&csi->sd);
> > + media_entity_cleanup(&csi->sd.entity);
> > +}
> > +
> > +/* Subdev notifier register */
> > +static int c3_mipi_csi_notify_bound(struct v4l2_async_notifier *notifier,
> > + struct v4l2_subdev *sd,
> > + struct v4l2_async_connection *asc)
> > +{
> > + struct csi_device *csi = v4l2_get_subdevdata(notifier->sd);
> > + struct media_pad *sink = &csi->sd.entity.pads[MIPI_CSI2_PAD_SINK];
> > + int ret;
> > +
> > + ret = media_entity_get_fwnode_pad(&sd->entity,
> > + sd->fwnode, MEDIA_PAD_FL_SOURCE);
> > + if (ret < 0) {
> > + dev_err(csi->dev, "Failed to find pad for %s\n", sd->name);
> > + return ret;
> > + }
> > +
> > + csi->src_sd = sd;
> > + csi->src_sd_pad = ret;
> > +
> > + return v4l2_create_fwnode_links_to_pad(sd, sink, MEDIA_LNK_FL_ENABLED |
> > + MEDIA_LNK_FL_IMMUTABLE);
> > +}
> > +
> > +static const struct v4l2_async_notifier_operations c3_mipi_csi_notify_ops = {
> > + .bound = c3_mipi_csi_notify_bound,
> > +};
> > +
> > +static int c3_mipi_csi_async_register(struct csi_device *csi)
> > +{
> > + struct v4l2_fwnode_endpoint vep = {
> > + .bus_type = V4L2_MBUS_CSI2_DPHY,
> > + };
> > + struct v4l2_async_connection *asc;
> > + struct fwnode_handle *ep;
> > + int ret;
> > +
> > + v4l2_async_subdev_nf_init(&csi->notifier, &csi->sd);
> > +
> > + ep = fwnode_graph_get_endpoint_by_id(dev_fwnode(csi->dev), 0, 0,
> > + FWNODE_GRAPH_ENDPOINT_NEXT);
> > + if (!ep)
> > + return -ENOTCONN;
> > +
> > + ret = v4l2_fwnode_endpoint_parse(ep, &vep);
> > + if (ret)
> > + goto err_put_handle;
> > +
> > + csi->bus = vep.bus.mipi_csi2;
> > + if (csi->bus.num_data_lanes != 2 && csi->bus.num_data_lanes != 4)
> > + goto err_put_handle;
>
> I would dev_err() here
>
> Thanks
> j
>
> > +
> > + asc = v4l2_async_nf_add_fwnode_remote(&csi->notifier, ep,
> > + struct v4l2_async_connection);
> > + if (IS_ERR(asc)) {
> > + ret = PTR_ERR(asc);
> > + goto err_put_handle;
> > + }
> > +
> > + csi->notifier.ops = &c3_mipi_csi_notify_ops;
> > + ret = v4l2_async_nf_register(&csi->notifier);
> > + if (ret)
> > + goto err_cleanup_nf;
> > +
> > + ret = v4l2_async_register_subdev(&csi->sd);
> > + if (ret)
> > + goto err_unregister_nf;
> > +
> > + fwnode_handle_put(ep);
> > +
> > + return 0;
> > +
> > +err_unregister_nf:
> > + v4l2_async_nf_unregister(&csi->notifier);
> > +err_cleanup_nf:
> > + v4l2_async_nf_cleanup(&csi->notifier);
> > +err_put_handle:
> > + fwnode_handle_put(ep);
> > + return ret;
> > +}
> > +
> > +static void c3_mipi_csi_async_unregister(struct csi_device *csi)
> > +{
> > + v4l2_async_unregister_subdev(&csi->sd);
> > + v4l2_async_nf_unregister(&csi->notifier);
> > + v4l2_async_nf_cleanup(&csi->notifier);
> > +}
> > +
> > +static int c3_mipi_csi_ioremap_resource(struct csi_device *csi)
> > +{
> > + struct device *dev = csi->dev;
> > + struct platform_device *pdev = to_platform_device(dev);
> > +
> > + csi->aphy = devm_platform_ioremap_resource_byname(pdev, "aphy");
> > + if (IS_ERR(csi->aphy))
> > + return PTR_ERR(csi->aphy);
> > +
> > + csi->dphy = devm_platform_ioremap_resource_byname(pdev, "dphy");
> > + if (IS_ERR(csi->dphy))
> > + return PTR_ERR(csi->dphy);
> > +
> > + csi->host = devm_platform_ioremap_resource_byname(pdev, "host");
> > + if (IS_ERR(csi->host))
> > + return PTR_ERR(csi->host);
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_configure_clocks(struct csi_device *csi)
> > +{
> > + const struct csi_info *info = csi->info;
> > + int ret;
> > + u32 i;
> > +
> > + for (i = 0; i < info->clock_num; i++)
> > + csi->clks[i].id = info->clocks[i];
> > +
> > + ret = devm_clk_bulk_get(csi->dev, info->clock_num, csi->clks);
> > + if (ret)
> > + return ret;
> > +
> > + for (i = 0; i < info->clock_num; i++) {
> > + if (!info->clock_rates[i])
> > + continue;
> > + ret = clk_set_rate(csi->clks[i].clk, info->clock_rates[i]);
> > + if (ret) {
> > + dev_err(csi->dev, "Failed to set %s rate %u\n", info->clocks[i],
> > + info->clock_rates[i]);
> > + return ret;
> > + }
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +static int c3_mipi_csi_probe(struct platform_device *pdev)
> > +{
> > + struct device *dev = &pdev->dev;
> > + struct csi_device *csi;
> > + int ret;
> > +
> > + csi = devm_kzalloc(dev, sizeof(*csi), GFP_KERNEL);
> > + if (!csi)
> > + return -ENOMEM;
> > +
> > + csi->info = of_device_get_match_data(dev);
> > + csi->dev = dev;
> > +
> > + ret = c3_mipi_csi_ioremap_resource(csi);
> > + if (ret)
> > + return dev_err_probe(dev, ret, "Failed to ioremap resource\n");
> > +
> > + ret = c3_mipi_csi_configure_clocks(csi);
> > + if (ret)
> > + return dev_err_probe(dev, ret, "Failed to configure clocks\n");
> > +
> > + platform_set_drvdata(pdev, csi);
> > +
> > + mutex_init(&csi->lock);
> > + pm_runtime_enable(dev);
> > +
> > + ret = c3_mipi_csi_subdev_init(csi);
> > + if (ret)
> > + goto err_disable_runtime_pm;
> > +
> > + ret = c3_mipi_csi_async_register(csi);
> > + if (ret)
> > + goto err_deinit_subdev;
> > +
> > + return 0;
> > +
> > +err_deinit_subdev:
> > + c3_mipi_csi_subdev_deinit(csi);
> > +err_disable_runtime_pm:
> > + pm_runtime_disable(dev);
> > + mutex_destroy(&csi->lock);
> > + return ret;
> > +};
> > +
> > +static void c3_mipi_csi_remove(struct platform_device *pdev)
> > +{
> > + struct csi_device *csi = platform_get_drvdata(pdev);
> > +
> > + c3_mipi_csi_async_unregister(csi);
> > + c3_mipi_csi_subdev_deinit(csi);
> > +
> > + pm_runtime_disable(&pdev->dev);
> > + mutex_destroy(&csi->lock);
> > +};
> > +
> > +static const struct csi_info c3_mipi_csi_info = {
> > + .clocks = {"vapb", "phy0"},
> > + .clock_rates = {0, 200000000},
> > + .clock_num = 2
> > +};
> > +
> > +static const struct of_device_id c3_mipi_csi_of_match[] = {
> > + { .compatible = "amlogic,c3-mipi-csi2",
> > + .data = &c3_mipi_csi_info,
> > + },
> > + { },
> > +};
> > +MODULE_DEVICE_TABLE(of, c3_mipi_csi_of_match);
> > +
> > +static struct platform_driver c3_mipi_csi_driver = {
> > + .probe = c3_mipi_csi_probe,
> > + .remove = c3_mipi_csi_remove,
> > + .driver = {
> > + .name = "c3-mipi-csi2",
> > + .of_match_table = c3_mipi_csi_of_match,
> > + .pm = &c3_mipi_csi_pm_ops,
> > + },
> > +};
> > +
> > +module_platform_driver(c3_mipi_csi_driver);
> > +
> > +MODULE_AUTHOR("Keke Li <keke.li@...ogic.com>");
> > +MODULE_DESCRIPTION("Amlogic C3 MIPI CSI-2 receiver");
> > +MODULE_LICENSE("GPL");
> >
> > --
> > 2.46.1
> >
> >
> >
Powered by blists - more mailing lists