@veye_xumm
mr123@ubuntu:~$ sudo dmesg |grep sc132
[sudo] password for mr123:
[ 11.962504] cs_sc132: module verification failed: signature and/or required key missing - tainting kernel
[ 11.964932] cssc132 9-003b: probing v4l2 sensor
[ 11.965743] cssc132 9-003b: tegracam sensor driver:cssc132_v2.0.6
[ 11.966346] cssc132 9-003b: read sensor id 0132
[ 11.966352] cssc132 9-003b: camera id is cs-mipi-sc132
[ 11.966405] tegra-camrtc-capture-vi tegra-capture-vi: subdev cssc132 9-003b bound
[ 11.970148] cssc132 9-003b: Detected CS_MIPI_SC132 camera
[ 11.977337] cssc132 10-003b: probing v4l2 sensor
[ 11.978659] cssc132 10-003b: tegracam sensor driver:cssc132_v2.0.6
[ 11.983230] cssc132 10-003b: cs_sc132_board_setup: error during i2c read probe (-121)
[ 11.983237] cssc132 10-003b: board setup failed
[ 11.983316] cssc132: probe of 10-003b failed with error -121
mr123@ubuntu:~$

@veye_xumm 我说的原厂系统指的就是nvidia原厂系统，然后驱动用的是你们预编译的ko和dtbo，只做了这些操作

另外我看wiki上面确实写了6.2.2搭配orin nano搭配cssc132相机确实是可以用的对吧，我只在昨天上午成功过一段时间，一直到现在都没有再成功过，我只能说我是完全按照wiki的步骤来的，这个现象太奇怪了，而且结合这个报错，有没有可能是我使用的fpc排线或者是开发板的差异，导致mipi信号畸变导致无法解析，如果可能的话能讲一下你们所使用的测试环境吗

@veye_xumm 我昨天测试了一天现象很奇怪，我第一次重刷成原厂系统后确实出图了，然后我又尝试的自编译驱动，也出图了，然后我逐步替换自编译的ko，然后替换到某个驱动的时候突然不行了，于是我回退开始确认这个现象，但是发现回退完还是不出图，然后我一路回退，回退到了预编译ko和设备树，发现也不行了！
更离谱的是我再次重刷，然后按早上的步骤重刷系统和预编译驱动，还是不行！
我现在想确认一下就是通过i2c配置的寄存器他是否会保存，断电后依然存在，是否可能是某个配置写死后没有清除导致的不恢复？

我今天会测试把nvidia的系统一烧进去，其他什么操作都不做就把git上预编译的ko和设备树按流程放进去来测试一下，测试完我也会同步进展，期待您的指导

@veye_xumm 以上是我的相机照片，您确认一下

如题，我尝试在orin nano上测试cssc132相机，我第一次使用的是在我的内核中移植驱动的方式来操作的，但是取流失败，在内核中打印如下[sudo] password for mr123:
mr123@mr123-desktop:~$ timeout 10s v4l2-ctl -d /dev/video0
--stream-mmap=4
--stream-count=30
--stream-to=/dev/null
--verbose
VIDIOC_QUERYCAP: ok
VIDIOC_REQBUFS returned 0 (Success)
VIDIOC_QUERYBUF returned 0 (Success)
VIDIOC_QUERYBUF returned 0 (Success)
VIDIOC_QUERYBUF returned 0 (Success)
VIDIOC_QUERYBUF returned 0 (Success)
VIDIOC_QBUF returned 0 (Success)
VIDIOC_QBUF returned 0 (Success)
VIDIOC_QBUF returned 0 (Success)
VIDIOC_QBUF returned 0 (Success)
VIDIOC_STREAMON returned 0 (Success)
mr123@mr123-desktop:~$ sudo dmesg -wT | grep -E 'cssc132|tegra-capture|nvcsi|rtcpu'
[六 8月 1 17:38:50 2026] tegra-camrtc-capture-vi tegra-capture-vi: uncorr_err: request timed out after 2500 ms
[六 8月 1 17:38:53 2026] tegra-camrtc-capture-vi tegra-capture-vi: uncorr_err: request timed out after 2500 ms
[六 8月 1 17:38:55 2026] tegra-camrtc-capture-vi tegra-capture-vi: uncorr_err: request timed out after 2500 ms
[六 8月 1 17:38:58 2026] tegra-camrtc-capture-vi tegra-capture-vi: uncorr_err: request timed out after 2500 ms
^C
mr123@mr123-desktop:~$
然后我换用预编译的驱动和设备树来测试，发现现象一致，其他信息，我用的版本是r36.5

@veye_xumm said in 关于第三方载板移植SC132M时发现只有当两个相机同时接入时才能工作的问题:

@robot

@robot said in 关于第三方载板移植SC132M时发现只有当两个相机同时接入时才能工作的问题:

如果方便的话，我是否需要上传我的设备树文件？希望能得到您的帮助

设备树文件要跟你的硬件载板想匹配。 具体来说，对每一个MIPI CSI-2接口，其i2c bus和mipi总线的配置，要匹配。

按照原理图来看是匹配的，我来对调mipi总线测试一下

@veye_xumm 是的，就是这个意思

@robot 事实上我现在对设备树进行了修改，去掉了另一路相机只保留了一路，然而他的现象又改变了：
[ 108.789772] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 108.796934] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 108.808227] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000089
[ 108.817041] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000089
[ 108.890376] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 108.998942] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.031040] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 109.038195] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 109.049455] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000089
[ 109.058265] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000089
[ 109.107458] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.215965] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.272185] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 109.279342] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 109.290629] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000089
[ 109.299435] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000089
[ 109.324479] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.432986] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.513296] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 109.520497] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 109.531754] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000089
[ 109.540546] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000089
[ 109.541482] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.645976] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.754488] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
[ 109.758441] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 109.758444] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 109.758675] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000089
[ 109.758679] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000089
[ 109.899159] tegra-i2c c250000.i2c: arb lost in communicate to add 0x32
如果方便的话，我是否需要上传我的设备树文件？希望能得到您的帮助

@robot 现象目前有所变化，dmesg打印变成了：
[ 20.658432] nvcsi 150c0000.nvcsi: csi4_stream_check_status (3) ERROR_STATUS2VI_VC0 = 0x00000004
[ 20.668079] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000199
[ 20.677014] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000199
[ 20.887656] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 20.894977] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 20.906432] nvcsi 150c0000.nvcsi: csi4_stream_check_status (3) ERROR_STATUS2VI_VC0 = 0x00000004
[ 20.916063] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000199
[ 20.924996] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000199
[ 21.135664] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 21.142981] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 21.154418] nvcsi 150c0000.nvcsi: csi4_stream_check_status (3) ERROR_STATUS2VI_VC0 = 0x00000004
[ 21.164068] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000199
[ 21.173013] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000199
[ 21.383666] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 21.390973] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 21.402430] nvcsi 150c0000.nvcsi: csi4_stream_check_status (3) ERROR_STATUS2VI_VC0 = 0x00000004
[ 21.412093] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_INTR_STATUS 0x00000199
[ 21.421037] nvcsi 150c0000.nvcsi: csi4_cil_check_status (3) CILA_ERR_INTR_STATUS 0x00000199
[ 21.631670] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 21.638990] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel

@veye_xumm d但是此时相机读取到的图像是全黑的，正常应该是实际拍到的东西

@veye_xumm said in 关于第三方载板移植SC132M时发现只有当两个相机同时接入时才能工作的问题:

如果检测不到，用

可以检测到相机，在/dev下有video0显示

rt，我们的载板使用的csi通道为2，3和标准开发板不同(0,2)，因此在使用RAW_MIPI_SC132M时，我们对设备树进行了修改，修改后成功的到了两个相机的图像
但是最近偶然发现当拔掉其中一只相机后，另一只也不能出图了，看了一下dmesg，有报错，打印如下：
[ 49.719292] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 49.727284] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 49.738728] nvcsi 150c0000.nvcsi: csi4_stream_check_status (2) ERROR_STATUS2VI_VC0 = 0x0000000c
[ 49.749641] nvcsi 150c0000.nvcsi: csi4_stream_check_status (2) ERROR_STATUS2VI_VC0 = 0x0000000c
在对照了半天标准设备树后未果，因此想请教一下，这可能是由什么原因导致的？

@veye_xumm 是的，确实是未装镜头导致的，实际图像正常，感谢您的指点

我们正在计划在自行研发的Tx2Nx控制器上面使用RAW_MIPI_SC132M这个模块，目前在调试阶段已经调到可以出图了，但是出图不对，看图像数据也都是00和FF，不像是正常图像
这个是我们用v4l2命令导出后用vooya打开的图像，我们不太清楚是哪里的问题，想得到您的指点

@veye_xumm said in Jeston Tx2 Nx + cs132 v4l2获取图像，一段时间后显示VIDIOC_DQBUF failed!Input/output error:

@robot 请问摄像机采用的什么工作模式？
另外请检查一下ffc排线是否插牢，左右对称。

手动曝光模式，1280*720P的模式下,另外我使用的是CS-FPD-CAM-SC132，且链接了两路，其他没有什么特殊的。
就是正常在抓流，代码也是根据教程来写的，如果您愿意帮忙排查也可以附上。
一开始也获取到了两个相机的图像，后来就有一路报错了，而且重新跑程序又可以获取到

以下为dmesg打印
[ 3210.118289] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERROR_STATUS2VI_VC2 = 0x00000006
[ 3210.126990] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERROR_STATUS2VI_VC2 = 0x00000002
[ 3210.135686] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) INTR_STATUS 0x00000008
[ 3210.143524] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERR_INTR_STATUS 0x00000008
[ 3274.243719] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 3274.250118] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 3274.259642] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERROR_STATUS2VI_VC2 = 0x00000006
[ 3274.268355] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERROR_STATUS2VI_VC2 = 0x00000002
[ 3289.542693] tegra-vi4 15700000.vi: Status: 2 channel:01 frame:0000
[ 3289.548980] tegra-vi4 15700000.vi: timestamp sof 3300200150752 eof 3300216795168 data 0x000000a0
[ 3289.558199] tegra-vi4 15700000.vi: capture_id 6758 stream 0 vchan 0
[ 3289.751480] tegra-vi4 15700000.vi: PXL_SOF syncpt timeout! err = -11
[ 3289.757887] tegra-vi4 15700000.vi: tegra_channel_error_recovery: attempting to reset the capture channel
[ 3289.767506] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERROR_STATUS2VI_VC2 = 0x00000006
[ 3289.776264] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERROR_STATUS2VI_VC2 = 0x00000002
[ 3289.785164] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) INTR_STATUS 0x00000004
[ 3289.793493] nvcsi 150c0000.nvcsi: csi4_stream_check_status (0) ERR_INTR_STATUS 0x00000004
请问如上打印是什么意思，我该按什么方向排查

@veye_xumm
好的，可以读到，那就没问题了，谢谢！

@veye_xumm
root@mr123-emmc3:/home/share/huanglielie/nvidia_jetson_veye_bsp-master/i2c_cmd/bin# ./cs_mipi_i2c.sh -w -f ledstrobe -p1 1
w led strobe enable
root@mr123-emmc3:/home/share/huanglielie/nvidia_jetson_veye_bsp-master/i2c_cmd/bin# ./cs_mipi_i2c.sh -r -f ledstrobe
NOT SUPPORTED!
您好，我测试发现返回了NOT SUPPORTED!
但是按照手册应该是支持这一功能的吧？
另外您提到的-b命令的用法是什么样的，我添加了-b 7 到命令里面，但是失败了
root@mr123-emmc3:/home/share/huanglielie/nvidia_jetson_veye_bsp-master/i2c_cmd/bin# ./cs_mipi_i2c.sh -r -f ledstrobe
NOT SUPPORTED!
root@mr123-emmc3:/home/share/huanglielie/nvidia_jetson_veye_bsp-master/i2c_cmd/bin# ./cs_mipi_i2c.sh -w -f -b 7 ledstrobe -p1 1
NOT SUPPORTED!