fatal error: nvscibuf.h: No such file or directory

[WangFengtu1996]

• Env

  (base) orin@orin-root:~/workspace/cuDLA-samples$ sudo jetson_release
  [sudo] password for orin:
  Software part of jetson-stats 4.2.4 - (c) 2024, Raffaello Bonghi
  Model: Jetson AGX Orin Developer Kit - Jetpack 5.1.2 [L4T 35.4.1]
  NV Power Mode[0]: MAXN
  Serial Number: [XXX Show with: jetson_release -s XXX]
  Hardware:
  - P-Number: p3701-0005
  - Module: NVIDIA Jetson AGX Orin (64GB ram)
  Platform:
  - Distribution: Ubuntu 20.04 focal
  - Release: 5.10.120-tegra
  jtop:
  - Version: 4.2.4
  - Service: Active
  Libraries:
  - CUDA: 11.4.315
  - cuDNN: 8.6.0.166
  - TensorRT: 5.1.2
  - VPI: 2.3.9
  - Vulkan: 1.3.204
  - OpenCV: 4.6.0 - with CUDA: YES

• build command I try both in hybrid mode and standalone mode. I got same error. src/cudla_context_standalone.h:30:10: fatal error: nvscibuf.h: No such file or directory

  make run 
  make run USE_DLA_STANDALONE_MODE=1

• I search solution in Internel. But I can not fix the bug . https://forums.developer.nvidia.com/t/where-is-nvscibuf-h/107802

[2yjia]

Same problem, have you solved it?

[WangFengtu1996]

NO. I think may need update jetpack. I want to use DLA HW by TensorRT. I read the code. It is not using TensorRT to inference model.

[2yjia]

I solved this problem. Refer to this URL：https://forums.developer.nvidia.com/t/where-are-the-nvsci-headers/221091 Download the BSP corresponding to jetpack. The corresponding header files contained in public_sources.tbz2. Modify the makefile and add the path

[WangFengtu1996]

About the library nvscibuf nvscisync, can modify Makefile,add library path

-L /usr/lib/aarch64-linux-gnu/tegra -l nvscibuf -l nvscisync