NVIDIA Thor support

[stama-hexagon]

Dear meta-tegra devs,

Thank you for the amazing work on the OE4T project.
Is there already a timeline for the meta-tegra support for the upcoming NVIDIA Thor Jetson?

[madisongh]

Not currently, no. Once NVIDIA releases a JetPack to support the new hardware, we can start planning for it.

[0xAl3xH]

is there any update on this?

[madisongh]

Well, I see NVIDIA just announced JetPack 7.0/L4T R38.2 for Thor five minutes ago... now that it's out, we can work on getting a branch together for it. We'll be somewhat hampered by lack of hardware, so if there is anyone who has a Thor unit and is willing to help with this work, please let me know.

[ichergui]

Hi @stama-hexagon @0xAl3xH
@madisongh and I will work on getting the Thor support in meta-tegra.
Do you mind if I ask a quick question.
Is this a request from company ? if so, which one ? or both individuals ?
Thanks

[stama-hexagon]

Thanks for getting back to this thread and I'm glad to hear your update. I'm working for hexagon and we're using meta-tegra in industrial context.

[minicarpet]

Are you working for this company ?

[ichergui]

Hi @stama-hexagon @0xAl3xH
There is some progress on meta-tegra with thor support.
Kindly see below:

• Linux basic command line

root@jetson-agx-thor-devkit:~# hostnamectl 
 Static hostname: jetson-agx-thor-devkit
       Icon name: computer
      Machine ID: 9a4a88a42bc84dacb45fe39bf42dd280
         Boot ID: 7a9778ed8f144cf3a99b1b1618ad47b9
    Product UUID: 802d7a23-e35c-50d8-9d86-e3d0ddfc483e
Operating System: OE4Tegra Demonstration Distro 5.2+snapshot-e7423ae09e62a1f5f72391e61cad2381f917c113 (master)
     CPE OS Name: cpe:/o:openembedded:tegrademo:5.2+snapshot-e7423ae09e62a1f5f72391e61cad2381f917c113
          Kernel: Linux 6.8.12-l4t-r38.2-g2e4cbbd7b538
    Architecture: arm64
 Hardware Vendor: NVIDIA
  Hardware Model: NVIDIA Jetson AGX Thor Developer Kit
 Hardware Serial: 1411625100021
Firmware Version: 38.2.0-gcid-41844464
   Firmware Date: Fri 2025-08-22
root@jetson-agx-thor-devkit:~#

• CUDA samples

root@jetson-agx-thor-devkit:~# /usr/bin/cuda-samples/deviceQuery 
/usr/bin/cuda-samples/deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "NVIDIA Thor"
  CUDA Driver Version / Runtime Version          13.0 / 13.0
  CUDA Capability Major/Minor version number:    11.0
  Total amount of global memory:                 125773 MBytes (131882532864 bytes)
  (020) Multiprocessors, (128) CUDA Cores/MP:    2560 CUDA Cores
  GPU Max Clock rate:                            1049 MHz (1.05 GHz)
  Memory Clock rate:                             0 Mhz
  Memory Bus Width:                              0-bit
  L2 Cache Size:                                 33554432 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
  Maximum Layered 1D Texture Size, (num) layers  1D=(32768), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(32768, 32768), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total shared memory per multiprocessor:        233472 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  1536
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 1 copy engine(s)
  Run time limit on kernels:                     Yes
  Integrated GPU sharing Host Memory:            Yes
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device supports Managed Memory:                Yes
  Device supports Compute Preemption:            Yes
  Supports Cooperative Kernel Launch:            Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 13.0, CUDA Runtime Version = 13.0, NumDevs = 1
Result = PASS
root@jetson-agx-thor-devkit:~#

@madisongh and I are working on the thor support. We will share it with everyone soon.