Lissanro

ik_llama.cpp debug log, part 2

May 23rd, 2025
67
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 5.84 KB | None | 0 0
  1. INFO [ launch_slot_with_task] slot is processing task | tid="137034227900416" timestamp=1747986998 id_slot=0 id_task=1055
  2. INFO [ update_slots] kv cache rm [p0, end) | tid="137034227900416" timestamp=1747986998 id_slot=0 id_task=1055 p0=708
  3. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 1 without access enabled
  4. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 1 without access enabled
  5. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 1 without access enabled
  6. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 1 without access enabled
  7. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 1 without access enabled
  8. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 1 without access enabled
  9. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 1 to device 0 without access enabled
  10. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 1 to device 0 without access enabled
  11. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 2 without access enabled
  12. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 2 without access enabled
  13. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 2 without access enabled
  14. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 2 without access enabled
  15. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 2 without access enabled
  16. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 2 without access enabled
  17. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 2 to device 0 without access enabled
  18. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 2 to device 0 without access enabled
  19. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 3 without access enabled
  20. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  21. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 3 without access enabled
  22. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  23. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 3 without access enabled
  24. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  25. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 3 to device 0 without access enabled
  26. ...
  27. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  28. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 3 without access enabled
  29. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  30. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 3 without access enabled
  31. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  32. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 3 to device 0 without access enabled
  33. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 3 to device 0 without access enabled
  34. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 1 without access enabled
  35. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 1 without access enabled
  36. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 1 without access enabled
  37. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 1 without access enabled
  38. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 1 to device 2 without access enabled
  39. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 1 to device 2 without access enabled
  40. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 2 without access enabled
  41. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 2 without access enabled
  42. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 2 to device 3 without access enabled
  43. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 2 to device 3 without access enabled
  44. ggml_backend_cuda_cpy_tensor_async: attempt to copy from device 0 to device 3 without access enabled
  45. ggml_backend_cuda_buffer_cpy_tensor: attempt to copy from device 0 to device 3 without access enabled
  46. INFO [ print_timings] prompt eval time = 529.80 ms / 2 tokens ( 264.90 ms per token, 3.77 tokens per second) | tid="137034227900416" timestamp=1747987209 id_slot=0 id_task=1055 t_prompt_processing=529.805 n_prompt_tokens_processed=2 t_token=264.9025 n_tokens_second=3.7749738111191857
  47. INFO [ print_timings] generation eval time = 209948.67 ms / 1528 runs ( 137.40 ms per token, 7.28 tokens per second) | tid="137034227900416" timestamp=1747987209 id_slot=0 id_task=1055 t_token_generation=209948.667 n_decoded=1528 t_token=137.400960078534 n_tokens_second=7.277969523855086
  48. INFO [ print_timings] total time = 210478.47 ms | tid="137034227900416" timestamp=1747987209 id_slot=0 id_task=1055 t_prompt_processing=529.805 t_token_generation=209948.667 t_total=210478.47199999998
  49. INFO [ update_slots] slot released | tid="137034227900416" timestamp=1747987209 id_slot=0 id_task=1055 n_ctx=102400 n_past=2237 n_system_tokens=0 n_cache_tokens=2237 truncated=false
  50. INFO [ update_slots] all slots are idle | tid="137034227900416" timestamp=1747987209
  51. INFO [ log_server_request] request | tid="136621299068928" timestamp=1747987209 remote_addr="127.0.0.1" remote_port=45008 status=200 method="POST" path="/completion" params={}
  52. INFO [ update_slots] all slots are idle | tid="137034227900416" timestamp=1747987209
Add Comment
Please, Sign In to add comment