Questions tagged [cubin]

3 questions
1
vote
1 answer

When should NVRTC compilation produce a CUBIN?

If I understand the workflow description in the NVRTC documentation correctly, here's how it works: Create an NVRTC program from the source text. Compile the NVRTC program to get PTX code. Device-link the PTX code using NVIDIA's Driver API…
einpoklum
  • 118,144
  • 57
  • 340
  • 684
1
vote
1 answer

Can I combine a "static" CUDA kernel launch with PTX code and get a working binary?

Suppose I take a CUDA program - for example the CUDA vectorAdd sample, and cut out the kernel's implementation, but still have the launch command: vectorAdd<<>>(d_A, d_B, d_C, numElements); and suppose that I write…
einpoklum
  • 118,144
  • 57
  • 340
  • 684
0
votes
1 answer

What's the unique id of a Cubin (cuda binary)

For common elf binaries, one could use readelf -n to get the unique build id as follows: $ readelf -n matrix Displaying notes found in: .note.gnu.build-id Owner Data size Description GNU …
flyingrose
  • 107
  • 2
  • 11