Nvmap error

Hello Nvidia

I’m having problems with accessing/detecting the gpu on orin-nx 16. Using jetpack6.2 .
#> ./deviceQuery
deviceQuery Starting…
CUDA Device Query (Runtime API) version (CUDART static linking)
NvRmMemInitNvmap failed with No such file or directory
356: Memory Manager Not supported
NvRmMemMgrInit failed error type: 196626
cudaGetDeviceCount returned 999
→ unknown error
Result = FAIL .

strace shows:

munmap(0xffff8ead5000, 43383) = 0
openat(AT_FDCWD, “/dev/nvmap”, O_RDWR|O_SYNC|O_CLOEXEC) = -1 ENOENT (No such file or directory)
write(2, "NvRmMemInitNvmap failed with No "…, 55NvRmMemInitNvmap failed with No such file or directory
) = 55

We see /dev/nvmap is missing. I checked nvmap & nvgpu modues:
nvmap loads on boot but doesn’t create /dev/nvmap (kernel 5.15) .
nvgpu doesn’t load on startup but loads manually with no errors and no meaningfull log on dmesg.

#> lsmod | grep nvgp
nvgpu 2654208 0
mc_utils 16384 2 nvgpu,tegra_camera_platform
host1x 180224 6 host1x_nvhost,host1x_fence,nvgpu,tegra_drm,nvhost_nvdla,nvhost_pva
nvmap 204800 1 nvgpu

#> lsmod | grep nvmap
nvmap 204800 1 nvgpu
nvsciipc 24576 1 nvmap

Our DeviceTree entries looks ok. The original dts of Jetpack6.2 also behave the same.

#> cat /proc/device-tree/bus@0/gpu@17000000/status
okay
#> cat /proc/device-tree/bus@0/gpu@17000000/compatible
nvidia,ga10b

#> ll /dev/nv*
crw-rw---- 1 root video 493, 0 Feb 20 13:24 /dev/nvhost-ctrl-isp
crw-rw---- 1 root video 497, 0 Feb 20 13:24 /dev/nvhost-ctrl-isp-thi
crw-rw---- 1 root video 494, 0 Feb 20 13:24 /dev/nvhost-ctrl-nvcsi
crw-rw---- 1 root video 501, 0 Feb 20 13:24 /dev/nvhost-ctrl-nvdla0
crw-rw---- 1 root video 499, 0 Feb 20 13:24 /dev/nvhost-ctrl-nvdla1
crw-rw---- 1 root video 491, 0 Feb 20 13:24 /dev/nvhost-ctrl-pva0
crw-rw---- 1 root video 495, 0 Feb 20 13:24 /dev/nvhost-ctrl-vi0
crw-rw---- 1 root video 500, 0 Feb 20 13:24 /dev/nvhost-ctrl-vi0-thi
crw-rw---- 1 root video 492, 0 Feb 20 13:24 /dev/nvhost-ctrl-vi1
crw-rw---- 1 root video 498, 0 Feb 20 13:24 /dev/nvhost-ctrl-vi1-thi
crw-rw---- 1 root video 489, 0 May 28 09:24 /dev/nvhost-power-gpu
crw-rw-rw- 1 root root 195, 255 Feb 20 13:26 /dev/nvidiactl
crw------- 1 root root 509, 0 Feb 20 13:24 /dev/nvme0
brw-rw---- 1 root disk 259, 0 Feb 20 13:24 /dev/nvme0n1
brw-rw---- 1 root disk 259, 1 Feb 20 13:24 /dev/nvme0n1p1
brw-rw---- 1 root disk 259, 2 Feb 20 13:24 /dev/nvme0n1p2
brw-rw---- 1 root disk 259, 3 Feb 20 13:24 /dev/nvme0n1p3
brw-rw---- 1 root disk 259, 4 Feb 20 13:24 /dev/nvme0n1p4
crw------- 1 root root 10, 124 Feb 20 13:26 /dev/nvme-fabrics
crw-rw-rw- 1 root root 507, 0 Feb 20 13:24 /dev/nvsciipc
drwxr-xr-x 2 root root 60 May 28 09:24 /dev/nvgpu/igpu0

#> tegrastats
05-28-2025 09:28:33 RAM 1032/15656MB (lfb 5x4MB) SWAP 0/7828MB (cached 0MB) CPU [0%@729,1%@729,0%@729,0%@729,0%@729,0%@729,0%@729,0%@729] EMC_FREQ 0%@2133 GR3D_FREQ 0%@[0] NVENC off NVDEC off NVJPG off NVJPG1 off VIC off OFA off NVDLA0 off NVDLA1 off PVA0_FREQ off APE 174 cv0@49.125C cpu@50.125C soc2@48.468C soc0@50.187C cv1@48.656C tj@53.156C soc1@53.156C cv2@48.906C VDD_IN
5298mW/5298mW VDD_CPU_GPU_CV 1099mW/1099mW VDD_SOC 1599mW/1599mW

Solving attempts so far:
I’m trying to build nvgpu & nvmap from Jetpack6.2 tegra-oot sources, but it has lots of compilation errors.
The Jetpack itself doesn’t build/use them. It uses already built modules from kernel_oot_modules.tbz2 .

Is the problem with nvmap not creating /dev/nvmap?
Is it a good method to build that modules from sources on the sdk?
Is there another solution?

Thanks.

Hi,

Do you meet this issue inside the container?
If so, please install nvidia-container-toolkit and launch the docker with --runtime=nvidia.

Thanks.

It’s not a docker. It’s a normal installation made by nvsdkmanager_flash.sh .

I tried playing with the devicetree even though it resembles kernel_tegra234-p3768-0000+p3767-0000-nv.dtb gpu entries.

I tried changing host1x, gpu@17000000, adding camera stuff we don’t need just in case from nv-soc/tegra234-soc-camera.dtsi .

Additional info:

#> cat /sys/kernel/debug/gpu.0/status
Channel Status - chip ga10b

PBDMA Status - chip ga10b

pbdma 0:
id: -1 - [channel] next_id: - -1 [channel] | status: invalid
PBDMA_PUT 000000dc5140e7d8 PBDMA_GET 00000018ee13cb24
GP_PUT 00000000 GP_GET 8daaed3b FETCH 00000000 HEADER e1d4f680
HDR d461114c SHADOW0 a2847415 SHADOW1 4cab7a13
pbdma 1:
id: -1 - [channel] next_id: - -1 [channel] | status: invalid
PBDMA_PUT 00000002e6f89f20 PBDMA_GET 000000b011dfb6e4
GP_PUT 00000000 GP_GET c8d6dacb FETCH 00000000 HEADER e1d1fe0c
HDR 613cd408 SHADOW0 f4ec7185 SHADOW1 93a2d385
pbdma 2:
id: -1 - [channel] next_id: - -1 [channel] | status: invalid
PBDMA_PUT 00000057f53665d4 PBDMA_GET 00000095166c6cf8
GP_PUT 00000000 GP_GET a947ad85 FETCH 00000000 HEADER c154f12c
HDR 01460cad SHADOW0 1333bd5a SHADOW1 01edb777
pbdma 3:
id: -1 - [channel] next_id: - -1 [channel] | status: invalid
PBDMA_PUT 00000045c3f9adc0 PBDMA_GET 0000001c5bb96144
GP_PUT 00000000 GP_GET 77357577 FETCH 00000000 HEADER e11616d0
HDR 68dbf9de SHADOW0 3652694b SHADOW1 a34c1394
pbdma 4:
id: -1 - [channel] next_id: - -1 [channel] | status: invalid
PBDMA_PUT 0000005fa87e80a8 PBDMA_GET 0000000000000000
GP_PUT 00000000 GP_GET 00000000 FETCH 00000000 HEADER 610234fc
HDR b57ff4cc SHADOW0 075813ae SHADOW1 66b0009a
pbdma 5:
id: -1 - [channel] next_id: - -1 [channel] | status: invalid
PBDMA_PUT 00000051e1c03f30 PBDMA_GET 000000e8998b03a8
GP_PUT 00000000 GP_GET 202a020d FETCH 00000000 HEADER c04505c4
HDR f233333c SHADOW0 00eada4e SHADOW1 34b89ab0
ga10b eng 0:
id: -1 (channel), next_id: -1 (channel), ctx status: invalid
ga10b eng 1:
id: -1 (channel), next_id: -1 (channel), ctx status: invalid
ga10b eng 2:
id: -1 (channel), next_id: -1 (channel), ctx status: invalid
ga10b eng 3:
id: -1 (channel), next_id: -1 (channel), ctx status: invalid
ga10b eng 4:
id: -1 (channel), next_id: -1 (channel), ctx status: invalid
ga10b eng 5:
id: -1 (channel), next_id: -1 (channel), ctx status: invalid

Doesn’t look good…

#> ls /sys/devices/platform/17000000.gpu/
aelpg_enable gpc0_tpc1_sm_lrf_ecc_single_err_count ltc0_lts2_tstg_ecc_parity_count
aelpg_param gpc0_tpc1_sm_rams_ecc_corrected_err_count ltc0_lts3_dstg_be_ecc_parity_count
allow_all gpc0_tpc1_sm_rams_ecc_uncorrected_err_count ltc0_lts3_ecc_ded_count
blcg_enable gpc0_tpc2_sm_cbu_ecc_corrected_err_count ltc0_lts3_ecc_sec_count
comptag_mem_deduct gpc0_tpc2_sm_cbu_ecc_uncorrected_err_count ltc0_lts3_rstg_ecc_parity_count
counters gpc0_tpc2_sm_icache_ecc_corrected_err_count ltc0_lts3_tstg_ecc_parity_count
counters_reset gpc0_tpc2_sm_icache_ecc_uncorrected_err_count mig_mode_config
devfreq gpc0_tpc2_sm_l1_data_ecc_corrected_err_count mig_mode_config_list
devfreq_dev gpc0_tpc2_sm_l1_data_ecc_uncorrected_err_count mig_mode_config_list_parsable
driver gpc0_tpc2_sm_l1_tag_ecc_corrected_err_count mmu_fillunit_ecc_corrected_err_count
driver_override gpc0_tpc2_sm_l1_tag_ecc_uncorrected_err_count mmu_fillunit_ecc_corrected_unique_err_count
elcg_enable gpc0_tpc2_sm_lrf_ecc_double_err_count mmu_fillunit_ecc_uncorrected_err_count
elpg_enable gpc0_tpc2_sm_lrf_ecc_single_err_count mmu_fillunit_ecc_uncorrected_unique_err_count
emc3d_ratio gpc0_tpc2_sm_rams_ecc_corrected_err_count mmu_hubtlb_ecc_corrected_err_count
enable_3d_scaling gpc0_tpc2_sm_rams_ecc_uncorrected_err_count mmu_hubtlb_ecc_corrected_unique_err_count
fbp_fs_mask gpc0_tpc3_sm_cbu_ecc_corrected_err_count mmu_hubtlb_ecc_uncorrected_err_count
fbp_pg_mask gpc0_tpc3_sm_cbu_ecc_uncorrected_err_count mmu_hubtlb_ecc_uncorrected_unique_err_count
flcg_enable gpc0_tpc3_sm_icache_ecc_corrected_err_count mmu_l2tlb_ecc_corrected_err_count
fmax_at_vmin_safe gpc0_tpc3_sm_icache_ecc_uncorrected_err_count mmu_l2tlb_ecc_corrected_unique_err_count
force_idle gpc0_tpc3_sm_l1_data_ecc_corrected_err_count mmu_l2tlb_ecc_uncorrected_err_count
freq_request gpc0_tpc3_sm_l1_data_ecc_uncorrected_err_count mmu_l2tlb_ecc_uncorrected_unique_err_count
golden_img_status gpc0_tpc3_sm_l1_tag_ecc_corrected_err_count modalias
gpc0_gcc_l15_ecc_corrected_err_count gpc0_tpc3_sm_l1_tag_ecc_uncorrected_err_count mscg_enable
gpc0_gcc_l15_ecc_uncorrected_err_count gpc0_tpc3_sm_lrf_ecc_double_err_count nvidia-gpu
gpc0_gpccs_ecc_corrected_err_count gpc0_tpc3_sm_lrf_ecc_single_err_count nvidia-gpu-power
gpc0_gpccs_ecc_uncorrected_err_count gpc0_tpc3_sm_rams_ecc_corrected_err_count nvidia-gpu-v2
gpc0_mmu_l1tlb_ecc_corrected_err_count gpc0_tpc3_sm_rams_ecc_uncorrected_err_count nvidia-gpu-v2-power
gpc0_mmu_l1tlb_ecc_uncorrected_err_count gpc_fs_mask of_node
gpc0_tpc0_sm_cbu_ecc_corrected_err_count gpc_pg_mask pmu_ecc_corrected_err_count
gpc0_tpc0_sm_cbu_ecc_uncorrected_err_count gpu_powered_on pmu_ecc_uncorrected_err_count
gpc0_tpc0_sm_icache_ecc_corrected_err_count gr0_fecs_ecc_corrected_err_count power
gpc0_tpc0_sm_icache_ecc_uncorrected_err_count gr0_fecs_ecc_uncorrected_err_count ptimer_ref_freq
gpc0_tpc0_sm_l1_data_ecc_corrected_err_count is_railgated ptimer_scale_factor
gpc0_tpc0_sm_l1_data_ecc_uncorrected_err_count ldiv_slowdown_factor ptimer_src_freq
gpc0_tpc0_sm_l1_tag_ecc_corrected_err_count load railgate_delay
gpc0_tpc0_sm_l1_tag_ecc_uncorrected_err_count ltc0_lts0_dstg_be_ecc_parity_count railgate_enable
gpc0_tpc0_sm_lrf_ecc_double_err_count ltc0_lts0_ecc_ded_count slcg_enable
gpc0_tpc0_sm_lrf_ecc_single_err_count ltc0_lts0_ecc_sec_count subsystem
gpc0_tpc0_sm_rams_ecc_corrected_err_count ltc0_lts0_rstg_ecc_parity_count supplier:platform:2c00000.memory-controller
gpc0_tpc0_sm_rams_ecc_uncorrected_err_count ltc0_lts0_tstg_ecc_parity_count supplier:platform:2c60000.external-memory-controller
gpc0_tpc1_sm_cbu_ecc_corrected_err_count ltc0_lts1_dstg_be_ecc_parity_count supplier:platform:bpmp
gpc0_tpc1_sm_cbu_ecc_uncorrected_err_count ltc0_lts1_ecc_ded_count tpc_fs_mask
gpc0_tpc1_sm_icache_ecc_corrected_err_count ltc0_lts1_ecc_sec_count tpc_pg_mask
gpc0_tpc1_sm_icache_ecc_uncorrected_err_count ltc0_lts1_rstg_ecc_parity_count tsg_timeslice_max_us
gpc0_tpc1_sm_l1_data_ecc_corrected_err_count ltc0_lts1_tstg_ecc_parity_count tsg_timeslice_min_us
gpc0_tpc1_sm_l1_data_ecc_uncorrected_err_count ltc0_lts2_dstg_be_ecc_parity_count uevent
gpc0_tpc1_sm_l1_tag_ecc_corrected_err_count ltc0_lts2_ecc_ded_count user
gpc0_tpc1_sm_l1_tag_ecc_uncorrected_err_count ltc0_lts2_ecc_sec_count
gpc0_tpc1_sm_lrf_ecc_double_err_count ltc0_lts2_rstg_ecc_parity_count

dmesg of nvgpu+dependencies load:

772.040804] tegra-host1x 13e00000.host1x: Adding to iommu group 8
[ 772.043626] host1x-context host1x-ctx.0: Adding to iommu group 9
[ 772.043944] host1x-context host1x-ctx.1: Adding to iommu group 10
[ 772.044087] host1x-context host1x-ctx.2: Adding to iommu group 11
[ 772.046374] host1x-context host1x-ctx.3: Adding to iommu group 12
[ 772.046760] host1x-context host1x-ctx.4: Adding to iommu group 13
[ 772.047668] host1x-context host1x-ctx.5: Adding to iommu group 14
[ 772.048273] host1x-context host1x-ctx.6: Adding to iommu group 15
[ 772.049432] host1x-context host1x-ctx.7: Adding to iommu group 16
[ 772.050665] host1x-context host1x-ctx.8: Adding to iommu group 17
[ 772.050915] host1x-context host1x-ctx.9: Adding to iommu group 18
[ 772.051130] host1x-context host1x-ctx.10: Adding to iommu group 19
[ 772.051497] host1x-context host1x-ctx.11: Adding to iommu group 20
[ 772.051818] host1x-context host1x-ctx.12: Adding to iommu group 21
[ 772.052423] host1x-context host1x-ctx.13: Adding to iommu group 22
[ 772.054681] host1x-context host1x-ctx.14: Adding to iommu group 23
[ 772.054990] host1x-context host1x-ctx.15: Adding to iommu group 24
[ 772.085567] mc: Linux media interface: v0.10
[ 772.085745] tegra-soc-hwpm f100000.hwpm: Adding to iommu group 25
[ 772.085981] nvsciipc: loaded module
[ 772.100531] nvmap_heap_init: nvmap_heap_init: created heap block cache
[ 772.102621] videodev: Linux video capture interface: v2.00
[ 772.103011] scare-pigeon 13e00000.host1x:vi0-thi@15f00000: Adding to iommu group 26
[ 772.118809] scare-pigeon 13e00000.host1x:vi0-thi@15f00000: syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 772.123751] scare-pigeon 13e00000.host1x:vi1-thi@14f00000: Adding to iommu group 27
[ 772.128530] pva 16000000.pva0: Adding to iommu group 28
[ 772.128605] tegra-vic 15340000.vic: Adding to iommu group 29
[ 772.133056] Inserting ivc_ext.ko module
[ 772.135733] scare-pigeon 13e00000.host1x:vi1-thi@14f00000: syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 772.136471] nvdla 15880000.nvdla0: Adding to iommu group 30
[ 772.140745] pva_iommu_context_dev 16��TCU debug prints will be routed to traces.
��000000.pva0:pva0_niso1_ctx0: Adding to iommu group 31
[ 772.140873] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx0: initialized (streamid=18, iommu=smmu.0x0000000008000000)
[ 772.140883] scare-pigeon 13e00000.host1x:isp-thi@14b00000: Adding to iommu group 32
[ 772.151944] scare-pigeon 13e00000.host1x:isp-thi@14b00000: syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 772.152162] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx1: Adding to iommu group 33
[ 772.152952] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx1: initialized (streamid=19, iommu=smmu.0x0000000008000000)
[ 772.153552] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx2: Adding to iommu group 34
[ 772.153658] pva_iommu_context_dev 16000000[ a0:772.213470_n7] tegra-ivc-bu: inits bc00000streamid=20,c-bus:mmu=smech0x0000vc 008nnel drig
5] pva_iommu_context_772dev 16000000.pva0:pva3479] isotx3teg: Ang ivcto iom-bumu c0000035.rtcpu772.154202] pva iv_iohanmmu_context_dev 16nel3
nitial772d (streamid=teg21, ioivc-bu=sms bc00000000008000000.rt):iv
us:dbg@2: iv464c chanpva drmmur mnteing
dev[ 772.213480003] .pva0:pva0_nisos b1_c: Addi000.rtto mu group c-bus:ivccontrol@3: ivc [ channel driver pvamis_iog
mmu[ ntext_.21 160004] 000.pvra-pva0_ns biso000tx4cpu: initialus:izecapture@4str: ivc cha22, ionne=smmu.river 000mis008000000)[ 772
7723485106] pvategivcmmu-bus bntedev000.rtcpu000.pvus:a0:pva0_niso1_cvc : Adding to ioml der mu group mis37g
772.155188] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx5: initialized (streamid=23, iommu=smmu.0x0000000008000000)
[ 772.155521] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx6: Adding to iommu group 38
[ 772.155605] pva_iommu_context_dev 16000000.pva0:pva7720_n5421_ctx69] LL devniticeial: fopsd (eamid=24, iommu=smmu.0x0000000008000000)
[ 772.155908] nctpva_ioion table_co alxt_devdy reg000.pva0:pva0_niso1_ctx7: Adding to iommu group 39d
[ 772.155987] pva_iommu_context_dev 16000000.pva0:pva0_niso1_ctx7: initialized (streamid=25, iommu=smmu.0x0000000008000000)
[ 772.156009] pva 16000000.pva0: pdata->version is HW_GEN2
[ 772.158559] nvdla 15880000.nvdla0: syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 772.170692] pva 16000000.pva0: Completed nvhost_client_device_init
[ 772.174470] nvdla 158c0000.nvdla1: Adding to iommu group 40
[ 772.182225] tegra-nvdec 15480000.nvdec: Adding to iommu group 41
[ 772.188696] nvdla 158c0000.nvdla1: syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 772.190345] pva 16000000.pva0: syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 10000
[ 772.197693] tegra186-cam-rtcpu bc00000.rtcpu: Adding to iommu group 42
[ 772.199547] tegra186-cam-rtcpu bc00000.rtcpu: Trace buffer configured at IOVA=0xbff00000
[ 772.200322] nvgpu: 17000000.gpu nvgpu_nvhost_syncpt_init:122 [INFO] syncpt_unit_base 60000000 syncpt_unit_size 4000000 size 100
[ 772.204670] tegra-nvenc 154c0000.nvenc: Adding to iommu group 43
[ 772.210828] tegra-ivc-bus bc00000.rtcpu:ivc-bus: region 0: iova=0xbfec0000-0xbfee027f size=131712
[ 772.211155] tegra-ivc-bus bc00000.rtcpu:ivc-bus:echo@0: echo: ver=0 grp=1 RX[16x64]=0x1000-0x1480 TX[16x64]=0x1480-0x1900
[ 772.212297] tegra-ivc-bus bc00000.rtcpu:ivc-bus:dbg@1: dbg: ver=0 grp=1 RX[1x512]=0x1900-0x1b80 TX[1x512]=0x1b80-0x1e00
[ 772.212589] tegra-ivc-bus bc00000.rtcpu:ivc-bus:dbg@2: dbg: ver=0 grp=1 RX[1x8192]=0x1e00-0x3e80 TX[1x8192]=0x3e80-0x5f00
[ 772.212623] tegra-ivc-bus bc00000.rtcpu:ivc-bus:ivccontrol@3: ivccontrol: ver=0 grp=1 RX[64x320]=0x5f00-0xaf80 TX[64x320]=0xaf80-0x10000
[ 772.212668] tegra-ivc-bus bc00000.rtcpu:ivc-bus:ivccapture@4: ivccapture: ver=0 grp=1 RX[512x64]=0x10000-0x18080 TX[512x64]=0x18080-0x2010
[ 772.212696] tegra-ivc-bus bc00000.rtcpu:ivc-bus:diag@5: diag: ver=0 grp=1 RX[1x64]=0x20100-0x201c0 TX[1x64]=0x201c0-0x20280
[ 772.212859] tegra186-cam-rtcpu bc00000.rtcpu: using cam RTCPU IRQ (216)
[ 772.212861] tegra186-cam-rtcpu bc00000.rtcpu: tegra_camrtc_mon_create is successful
[ 772.213117] tegra186-cam-rtcpu bc00000.rtcpu: firmware version cpu=rce cmd=6 sha1=e2238c99959d2df9350d393f04e1f44e5bef98cb
[ 772.213154] tegra-nvjpg 15380000.nvjpg: Adding to iommu group 44
[ 772.213477] tegra-ivc-bus bc00000.rtcpu:ivc-bus:echo@0: ivc channel driver missing
[ 772.213479] tegra-ivc-bus bc00000.rtcpu:ivc-bus:dbg@1: ivc channel driver missing
[ 772.213481] tegra-ivc-bus bc00000.rtcpu:ivc-bus:dbg@2: ivc channel driver missing
[ 772.213483] tegra-ivc-bus bc00000.rtcpu:ivc-bus:ivccontrol@3: ivc channel driver missing
[ 772.213484] tegra-ivc-bus bc00000.rtcpu:ivc-bus:ivccapture@4: ivc channel driver missing
[ 772.213486] tegra-ivc-bus bc00000.rtcpu:ivc-bus:diag@5: ivc channel driver missing
[ 772.231910] tegra-nvjpg 15540000.nvjpg: Adding to iommu group 45
[ 772.243748] tegra-ofa 15a50000.ofa: Adding to iommu group 46
[ 772.244815] drm drm: [drm] Cannot find any crtc or sizes
[ 772.245035] [drm] Initialized tegra 1.0.99 20120330 for drm on minor 0
[ 772.257650] [RCE] TCU debug prints will be routed to traces.
[ 772.302129] tegra194-isp5 14800000.isp: Adding to iommu group 32
[ 772.302192] tegra194-vi5 13e00000.host1x:vi0@15c00000: Adding to iommu group 26
[ 772.304437] driver not enabled, cannot register any devices
[ 772.305190] driver not enabled, cannot register any devices
[ 772.305324] tegra194-vi5 13e00000.host1x:vi1@14c00000: Adding to iommu group 27
[ 772.305429] (NULL device *): fops function table already registered
[ 772.305727] driver not enabled, cannot register any devices
[ 772.306642] driver not enabled, cannot register any devices

Hi,

Do you flash the board with JetPack 6.2?
Or have you applied any customization to the BSP?

Thanks.

Hello. I’m continuing where Eliyahu1 (colleague) stopped.
I did flash the board with JP6.2.
Also, we customized the BSP ofcourse. DT, kernel config and filesystem. But as I mentioned. It works the same when using the JP6.2 default kernel+modules & default dtb which is kernel_tegra234-p3768-0000+p3767-0000-nv.dtb

I think this part of dmesg is important:
[ 772.213486] tegra-ivc-bus bc00000.rtcpu:ivc-bus:diag@5: ivc channel driver missing
[ 772.231910] tegra-nvjpg 15540000.nvjpg: Adding to iommu group 45
[ 772.243748] tegra-ofa 15a50000.ofa: Adding to iommu group 46
[ 772.244815] drm drm: [drm] Cannot find any crtc or sizes
[ 772.245035] [drm] Initialized tegra 1.0.99 20120330 for drm on minor 0
[ 772.257650] [RCE] TCU debug prints will be routed to traces.
[ 772.302129] tegra194-isp5 14800000.isp: Adding to iommu group 32
[ 772.302192] tegra194-vi5 13e00000.host1x:vi0@15c00000: Adding to iommu group 26
[ 772.304437] driver not enabled, cannot register any devices
[ 772.305190] driver not enabled, cannot register any devices
[ 772.305324] tegra194-vi5 13e00000.host1x:vi1@14c00000: Adding to iommu group 27
[ 772.305429] (NULL device *): fops function table already registered
[ 772.305727] driver not enabled, cannot register any devices
[ 772.306642] driver not enabled, cannot register any devices

It’ll great If U can supply a newer version of:
kernel_supplements.tbz2
kernel_oot_modules.tbz2

for us to test with your newly configured kernel + oot modules & dtb .

thanks

Hi,
The latest release is Jetpack 6.2 r36.4.3. Yo can download the source code from

Jetson Linux | NVIDIA Developer
Driver Package (BSP) Sources

And check steps in Kernel Customization:

Kernel Customization — NVIDIA Jetson Linux Developer Guide

Let’s summary the problem:
“/dev/nvmap” is not created by nvmap , so cuda can’t see the gpu.
We’re using Jetpack6.2 , Jetson r36.4.3. Not running in docker.
nvmap & nvgpu modules load fine. But there’s a registeration error that might be related:
[ 772.257650] [RCE] TCU debug prints will be routed to traces.
[ 772.302129] tegra194-isp5 14800000.isp: Adding to iommu group 32
[ 772.302192] tegra194-vi5 13e00000.host1x:vi0@15c00000: Adding to iommu group 26
[ 772.304437] driver not enabled, cannot register any devices
[ 772.305190] driver not enabled, cannot register any devices
[ 772.305324] tegra194-vi5 13e00000.host1x:vi1@14c00000: Adding to iommu group 27
[ 772.305429] (NULL device *): fops function table already registered
[ 772.305727] driver not enabled, cannot register any devices
[ 772.306642] driver not enabled, cannot register any devices

maybe there’s a new BETA version or unreleased Jetpack6.3 version of nvmap nvgpu (kernel_oot_modules.tbz2) ?

Hi,
We don’t see the issue on our developer kit(Orin NX module + Orin Nano carrier board). Would suggest try on this setup and see if you observe the issue.

Hello
Can a carrier board effect the GPU? Logic says because it’s part of the soc, if it gets power it can’t be corrupted by the board. is it so?

Does the Orin nano (evaluation?) board fits Orin-nx? Cause we know Orin-nx doesn’t have evaluation board.

We have Auvidea JNX45 . If the gpu works there, will it be good enough for U to escalate the problem? or use Orin-nano carrier exactly?

thanks.