Thor display

Robotics & Edge Computing Jetson Systems Jetson Thor
1

因为做了新的内核移植，bsp 除了内核以外相关显示源码没有修改 现在文件系统里面opensrc-disp相关驱动加载异常

  1. sudo modprobe nvidia
  2. [ 1538.799130] cpufreq: cpu0,cur:1061000,set:2601000,delta:1540000,set ndiv:289
    [ 1538.819485] NVRM: The NVIDIA probe routine was not called for 2 device(s).
    [ 1538.819495] NVRM: This can occur when another driver was loaded and
    NVRM: obtained ownership of the NVIDIA device(s).
    [ 1538.819498] NVRM: Try unloading the conflicting kernel module (and/or
    NVRM: reconfigure your kernel without the conflicting
    NVRM: driver(s)), then try loading the NVIDIA kernel module
    NVRM: again.
    [ 1538.819500] NVRM: No NVIDIA devices probed.
    [ 1538.822273] nvidia-nvlink: Unregistered Nvlink Core, major device number 4
2

What is the result of lsmod now?

3

dmesg | tail 就是上面那个

4

我問的問題跟你的回覆好像沒有太相關…

5

就是没法加载上去 打印上面LOG

7

lsmod.log (5.7 KB)

8

請列一下你這幾個driver build的步驟

10

make -C kernel

make modules

sudo -E make install -C kernel

sudo -E make modules_install

11

請問你是在Jetson上編譯還是在x86 host?

12

x86 ununtu24

13

由於我們這邊沒辦法複製到你碰上的問題, 可能請你把完整步驟列出. 包含toolchain使用哪板的資訊也列出來

14

全是38.0 BSP 里面东西 除了内核不一样 ，驱动ko 都是有的 nvlink初始化失败了 导致后面不能加载

15

所有操作 按照文档流程 可以确定没问题的

18

@WayneWWW 我按照6.8.12版本内核去移植 驱动加载到读取寄存器失败

NVRM: gpuHandleSanityCheckRegReadError_GH100: Possible bad register read: addr: 0x110040, regvalue: 0xbadf5044, error code: NV_PPRIV_SYS_PRI_ERROR_CODE_FECS_PRI_CLIENT_ERR
[ 15.293063] NVRM: gpuHandleSanityCheckRegReadError_GH100: Possible bad register read: addr: 0x1100f4, regvalue: 0xbadf5044, error code: NV_PPRIV_SYS_PRI_ERROR_CODE_FECS_PRI_CLIENT_ERR
[ 15.293067] NVRM: gpuHandleSanityCheckRegReadError_GH100: Possible bad register read: addr: 0x110040, regvalue: 0xbadf5044, error code: NV_PPRIV_SYS_PRI_ERROR_CODE_FECS_PRI_CLIENT_ERR

19

NVRM: RmInitAdapter: Cannot initialize GSP firmware RM
[ 15.305999] NVRM: iovaspaceDestruct_IMPL: 1 left-over mappings in IOVAS 0x100
[ 15.306010] NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x62:0x55:1994)
[ 15.307793] NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
[ 15.349491] [RF]halrf_ic_cfg_init aid = 0x6758 !!
[ 15.349497] [RF]hal_i->chip_id = 3 !!
[ 15.349499] [RF]rf->ic_type = 0x8 !!
[ 15.415394] [drm:nv_drm_dev_load [nvidia_drm]] ERROR [nvidia-drm] [GPU ID 0x00000100] Failed to allocate NvKmsKapiDevice
[ 15.415431] [drm:nv_drm_register_drm_device [nvidia_drm]] ERROR [nvidia-drm] [GPU ID 0x00000100] Failed to load device
[ 15.423532] [drm] [nvidia-drm] [GPU ID 0x00020000] Loading driver
[ 15.427004] [WARNING][BB][halbb_cck_rpl_ofst_ref_init] There is no cck_rpl_ofst_ref !
[ 15.438167] [BB][halbb_la_bb_set_smp_rate] smp_rate_tmp=7, la_smp_rate_log=160 M
[ 15.438186] [BB]addr=[0x30000 ~ 0x70000], buf_size=(256 K), smp_num_max=(32768)
[ 15.694978] nvidia-modeset: WARNING: HW supports 8 heads. Limiting to 4 heads
[ 15.695421] nvidia-modeset: WARNING: GPU:0: The head configuration (0xff) is inconsistent with the number of heads (4)
[ 15.783755] NVRM: rpcRmApiControl_dce: NVRM_RPC_DCE: Failed RM ctrl call cmd:0x731341 result 0xffff: Failure: Generic Error [NV_ERR_GENERIC]
[ 16.061046] [drm] Initialized nvidia-drm 0.0.0 20160202 for 8808c00000.display on minor 1
[ 16.061055] [drm] [nvidia-drm] [GPU ID 0x00020000] Invalid framebuffer console info
[ 16.373571] Console: switching to colour frame buffer device 240x67
[ 16.379112] nv_platform 8808c00000.display: [drm] fb0: nvidia-drmdrmfb frame buffer device
[ 18.317038] nvidia 0000:01:00.0: Enabling HDA controller

20

@WayneWWW 如果有需要 我可以给源码上传git

21

We have tried to build kernel on 38.2 but could not reproduce this issue.