Guest User

Untitled

a guest
Aug 29th, 2021
99
267 days
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
  1. /opt/rocm/opencl/bin/clinfo
  2. Number of platforms: 1
  3. Platform Profile: FULL_PROFILE
  4. Platform Version: OpenCL 2.0 AMD-APP (3305.0)
  5. Platform Name: AMD Accelerated Parallel Processing
  6. Platform Vendor: Advanced Micro Devices, Inc.
  7. Platform Extensions: cl_khr_icd cl_amd_event_callback
  8.  
  9.  
  10. Platform Name: AMD Accelerated Parallel Processing
  11. Number of devices: 1
  12. Device Type: CL_DEVICE_TYPE_GPU
  13. Vendor ID: 1002h
  14. Board name: Ellesmere [Radeon RX 470/480/570/570X/580/580X/590]
  15. Device Topology: PCI[ B#1, D#0, F#0 ]
  16. Max compute units: 36
  17. Max work items dimensions: 3
  18. Max work items[0]: 1024
  19. Max work items[1]: 1024
  20. Max work items[2]: 1024
  21. Max work group size: 256
  22. Preferred vector width char: 4
  23. Preferred vector width short: 2
  24. Preferred vector width int: 1
  25. Preferred vector width long: 1
  26. Preferred vector width float: 1
  27. Preferred vector width double: 1
  28. Native vector width char: 4
  29. Native vector width short: 2
  30. Native vector width int: 1
  31. Native vector width long: 1
  32. Native vector width float: 1
  33. Native vector width double: 1
  34. Max clock frequency: 1360Mhz
  35. Address bits: 64
  36. Max memory allocation: 7301444400
  37. Image support: Yes
  38. Max number of images read arguments: 128
  39. Max number of images write arguments: 8
  40. Max image 2D width: 16384
  41. Max image 2D height: 16384
  42. Max image 3D width: 16384
  43. Max image 3D height: 16384
  44. Max image 3D depth: 8192
  45. Max samplers within kernel: 26591
  46. Max size of kernel argument: 1024
  47. Alignment (bits) of base address: 1024
  48. Minimum alignment (bytes) for any datatype: 128
  49. Single precision floating point capability
  50. Denorms: No
  51. Quiet NaNs: Yes
  52. Round to nearest even: Yes
  53. Round to zero: Yes
  54. Round to +ve and infinity: Yes
  55. IEEE754-2008 fused multiply-add: Yes
  56. Cache type: Read/Write
  57. Cache line size: 64
  58. Cache size: 16384
  59. Global memory size: 8589934592
  60. Constant buffer size: 7301444400
  61. Max number of constant args: 8
  62. Local memory type: Scratchpad
  63. Local memory size: 65536
  64. Max pipe arguments: 16
  65. Max pipe active reservations: 16
  66. Max pipe packet size: 3006477104
  67. Max global variable size: 7301444400
  68. Max global variable preferred total size: 8589934592
  69. Max read/write image args: 64
  70. Max on device events: 1024
  71. Queue on device max size: 8388608
  72. Max on device queues: 1
  73. Queue on device preferred size: 262144
  74. SVM capabilities:
  75. Coarse grain buffer: Yes
  76. Fine grain buffer: Yes
  77. Fine grain system: No
  78. Atomics: No
  79. Preferred platform atomic alignment: 0
  80. Preferred global atomic alignment: 0
  81. Preferred local atomic alignment: 0
  82. Kernel Preferred work group size multiple: 64
  83. Error correction support: 0
  84. Unified memory for Host and Device: 0
  85. Profiling timer resolution: 1
  86. Device endianess: Little
  87. Available: Yes
  88. Compiler available: Yes
  89. Execution capabilities:
  90. Execute OpenCL kernels: Yes
  91. Execute native function: No
  92. Queue on Host properties:
  93. Out-of-Order: No
  94. Profiling : Yes
  95. Queue on Device properties:
  96. Out-of-Order: Yes
  97. Profiling : Yes
  98. Platform ID: 0x7f8035741e10
  99. Name: gfx803
  100. Vendor: Advanced Micro Devices, Inc.
  101. Device OpenCL C version: OpenCL C 2.0
  102. Driver version: 3305.0 (HSA1.1,LC)
  103. Profile: FULL_PROFILE
  104. Version: OpenCL 1.2
  105. Extensions: cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_amd_device_attribute_query cl_amd_media_ops cl_amd_media_ops2 cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program
RAW Paste Data