使用GMSL的串行器和解串器9295 9296 两路imx390相机 有两种采集方式，一种是在前端拼接好作为一路采集； 另一种是独立采集，使用MIPI的虚拟通道； 下面为调试记录
2. 搭建环境设备树：
只截取部分 { num-channels = <2>; ports { #address-cells = <1>; #size-cells = <0>; port@0 { reg = <0>; imx390_vi_in0: endpoint { vc-id = <0>; port-index = <0>; bus-width = <2>; remote-endpoint = <&imx390_csi_out0>; }; }; port@1 { reg = <1>; imx390_vi_in1: endpoint { vc-id = <1>; port-index = <0>; bus-width = <2>; remote-endpoint = <&imx390_csi_out1>; }; }; }; } mode0 {/*mode IMX390_MODE_1920X1080_CROP_30FPS*/ mclk_khz = “24000”; num_lanes = “2”; tegra_sinterface = “serial_a”; vc_id = “0”; discontinuous_clk = “no”; dpcm_enable = “false”; cil_settletime = “0”; dynamic_pixel_bit_depth = “8”; csi_pixel_bit_depth = “8”; mode_type = “bayer”; pixel_phase = “rggb”; active_w = “256”; active_h = “1”; readout_orientation = “0”; line_length = “300”; inherent_gain = “1”; pix_clk_hz = “337500000”; serdes_pix_clk_hz = “337500000”; gain_factor = “10”; min_gain_val = “0”; /* dB */ max_gain_val = “300”; /* dB */ step_gain_val = “3”; /* 0.3 */ default_gain = “0”; min_hdr_ratio = “1”; max_hdr_ratio = “1”; framerate_factor = “1000000”; min_framerate = “30000000”; max_framerate = “30000000”; step_framerate = “1”; default_framerate = “30000000”; exposure_factor = “1000000”; min_exp_time = “59”; /*us, 2 lines*/ max_exp_time = “33333”; step_exp_time = “1”; default_exp_time = “33333”;/* us */ embedded_metadata_height = “0”; };驱动：使用官方驱动直接进行移植
/hardware/nvidia/platform/t19x/下的两个设备树文件夹。 jakku/是NX平台的DT来源，galen/包括Xavier平台的DT。
驱动加载ok
[ 5.334391] imx390 30-001b: probing v4l2 sensor. [ 5.335400] imx390 30-001b: tegracam sensor driver:imx390_v2.0.6 [ 5.335498] imx390 30-001b: Detected IMX390 sensor [ 5.335586] imx390 30-001c: probing v4l2 sensor. [ 5.335767] imx390 30-001c: couldn’t create debugfs [ 5.335970] imx390 30-001c: tegracam sensor driver:imx390_v2.0.6 [ 5.336033] imx390 30-001c: Detected IMX390 sensor 3. 视频采集成功注册了Xavier NX板中的imx390驱动程序。 当采集数据时产生错误
kworker/0:0-4 [000] …. 436.167754: rtos_queue_peek_from_isr_failed: tstamp:14032721488 queue:0x0bcbcf78 v4l2-ctl-10518 [001] …. 436.255902: tegra_channel_set_stream: enable : 0x0 v4l2-ctl-10518 [001] …. 436.255906: tegra_channel_set_stream: imx390 30-001c : 0x0 v4l2-ctl-10518 [001] …. 436.255918: tegra_channel_set_stream: 15a00000.nvcsi–1 : 0x0 v4l2-ctl-10518 [001] …. 436.255920: csi_s_stream: enable : 0x0 v4l2-ctl-10518 [000] …. 436.261468: tegra_channel_set_power: imx390 30-001c : 0x0 v4l2-ctl-10518 [000] …. 436.261483: camera_common_s_power: status : 0x0 v4l2-ctl-10518 [000] …. 436.261617: tegra_channel_set_power: 15a00000.nvcsi–1 : 0x0 v4l2-ctl-10518 [000] …. 436.261620: csi_s_power: enable : 0x0 kworker/0:0-4 [000] …. 436.279807: rtos_queue_send_from_isr_failed: tstamp:14036251427 queue:0x0bcb41f8 kworker/0:0-4 [000] …. 436.279823: rtos_queue_send_from_isr_failed: tstamp:14036251581 queue:0x0bcb8a60 kworker/0:0-4 [000] …. 436.279825: rtos_queue_send_from_isr_failed: tstamp:14036251737 queue:0x0bcba5e0 kworker/0:0-4 [000] …. 436.279826: rtos_queue_send_from_isr_failed: tstamp:14036251
（1）检查设备树的属性设置。 这些DT属性应该与实际的传感器MIPI信号相匹配。 VI引擎设置缓冲区接收信号根据设备树的属性
（2）看起来VI引擎没有收到帧信号。 看到PXL_SOF和PXL_EOF，表示帧的开始和结束。
（3）检查传感器像素时钟设置 相机软件使用传感器像素时钟来计算曝光和帧率的传感器。 必须正确配置pix_clk_hz
按照上面的思路，调整了设备树的各种参数 实现了2路拼接视频采集
4. 使用虚拟通道独立采集（1）参考多虚拟通道驱动
设备树属性添加VC -id和vc_id的VC支持。 参考如下；
$L4T_Sources/r32.4.3/Linux_for_Tegra/source/public/hardware/nvidia/platform/t18x/common/kernel-dts/t18x-common-modules/tegra186-camera-imx390-a00.dtsi
（2）调整端口绑定关系
启用了4个虚拟通道，驱动程序也成功加载。 但是sub dev和/dev/ videoo2和/dev/video3没有注册。我 日志如下
[ 0.000000] Inode-cache hash table entries: 524288 (order: 10, 4194304 bytes) [ 0.429603] DTS File Name: …/arch/arm64/boot/dts/…/…/…/…/…/…/hardware/nvidia/platform/t19x/jakku/kernel-dts/tegra194-p3668-all-p3509-0000.dts [ 0.700706] DTS File Name: …/arch/arm64/boot/dts/…/…/…/…/…/…/hardware/nvidia/platform/t19x/jakku/kernel-dts/tegra194-p3668-all-p3509-0000.dts [ 0.719456] iommu: Adding device 3180000.i2c to group 12 [ 0.788194] vdd-sdmmc1-sw: 3300 mV [ 1.457135] t194-nvcsi 15a00000.nvcsi: initialized [ 2.118946] tegra194_cpufreq_probe: platform driver Initialization: pass [ 2.146144] misc nvmap: cvsram :dma coherent mem declare 0x0000000050000000,4194304 [ 2.993281] tegra194-isp5 14800000.isp: initialized [ 2.999694] tegra194-vi5 15c10000.vi: using default number of vi channels, 36 [ 3.006886] tegra194-vi5 15c10000.vi: initialized [ 3.010860] tegra194-vi5 15c10000.vi: subdev 15a00000.nvcsi–2 bound [ 3.010929] tegra194-vi5 15c10000.vi: subdev 15a00000.nvcsi–1 bound [ 5.429971] tegra194-vi5 15c10000.vi: subdev imx390 30-001b bound [ 5.431616] tegra194-vi5 15c10000.vi: subdev imx390 30-001c bound [ 0.573907] CPU4: Booted secondary processor [4e0f0040] [ 5.429320] imx390 30-001b: probing v4l2 sensor. [ 5.429919] imx390 30-001b: tegracam sensor driver:imx390_v2.0.6 [ 5.429971] tegra194-vi5 15c10000.vi: subdev imx390 30-001b bound [ 5.431118] imx390 30-001b: Detected IMX390 sensor [ 5.431185] imx390 30-001c: probing v4l2 sensor. [ 5.431366] imx390 30-001c: couldn’t create debugfs [ 5.431579] imx390 30-001c: tegracam sensor driver:imx390_v2.0.6 [ 5.431616] tegra194-vi5 15c10000.vi: subdev imx390 30-001c bound [ 5.432205] imx390 30-001c: Detected IMX390 sensor [ 5.432288] imx390 30-001d: probing v4l2 sensor. [ 5.432466] imx390 30-001d: couldn’t create debugfs [ 5.432664] imx390 30-001d: tegracam sensor driver:imx390_v2.0.6 [ 5.432700] imx390 30-001d: Detected IMX390 sensor [ 5.432812] imx390 30-001e: probing v4l2 sensor. [ 5.432950] imx390 30-001e: couldn’t create debugfs [ 5.433124] imx390 30-001e: tegracam sensor driver:imx390_v2.0.6 [ 5.433154] imx390 30-001e: Detected IMX390 sensor调整设备树的端口绑定关系； 可解决上述问题
（3）MIPI速率问题
驱动程序在850Mbps以上工作正常。 但它不能在低于850Mbps的数据速率下工作。 我已经改变了解决时间，然后也没有改善
传感器可以通过配置将速率从200mbps配到1.3Gbps。 通过在imx390中配置传感器和dtsi(像素时钟和定位时间)文件来测试不同的数据速率。 在1.3Gbps和900Mbps的速度下工作，但是当传感器配置为650、450、220和350 Mbps时，它就不工作了
（4）添加set_mode_delay_ms
Xavier的VI引擎是定时敏感的， 尝试添加set_mode_delay_ms属性来增加第一帧的超时值。
set_mode_delay_ms |捕获开始后第一帧的最大等待时间，以毫秒为单位。修改如下：
调试后仍无变化
（5）尝试设置固定硬编码时间
$L4T_Sources/r32.4.4/Linux_for_Tegra/source/public/kernel/nvidia/drivers/media/platform/tegra/camera/nvcsi/csi5_fops.c
static int csi5_stream_set_config(struct tegra_csi_channel *chan, u32 stream_id, u32 csi_port, int csi_lanes) { … unsigned int cil_settletime = read_settle_time_from_dt(chan);修改后也无变化
5. 结尾目前在高速率下是可以正常采集的； 但是配置低速率无法工作问题暂时还没解决； 有待后续深究
