Advertisement
illwieckz

AMD Orca, PAL, ROCm: cl_amd_device_attribute_query

Jun 24th, 2023 (edited)
181
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 18.16 KB | None | 0 0
  1. ## GCN1, Orca (maybe wrong, the fact Orca lists them is likely a bug)
  2.  
  3. Device Name ATI FirePro V (FireGL V) Graphics Adapter (TAHITI, DRM 3.41.0, 5.13.0-39-generic, LLVM 9.0.1)
  4. Device Name ATI FirePro V (FireGL V) Graphics Adapter (TAHITI, DRM 3.41.0, 5.13.0-39-generic, LLVM 9.0.1)
  5. Platform Name AMD Accelerated Parallel Processing
  6. Device Name Tahiti
  7. Device Version OpenCL 1.2 AMD-APP (3224.4)
  8. Driver Version 3224.4
  9. Device Type GPU
  10. Device Board Name (AMD) ATI FirePro V (FireGL V) Graphics Adapter
  11. Device Topology (AMD) PCI-E, 03:00.0
  12. SIMD per compute unit (AMD) 4
  13. SIMD width (AMD) 16
  14. SIMD instruction width (AMD) 1
  15. Graphics IP (AMD) 6.0
  16. Preferred work group size (AMD) 256
  17. Max work group size (AMD) 1024
  18. Wavefront width (AMD) 64
  19. Global free memory (AMD) <printDeviceInfo:78: get number of CL_DEVICE_GLOBAL_FREE_MEMORY_AMD : error -33>
  20. Global memory channels (AMD) 12
  21. Global memory banks per channel (AMD) 16
  22. Global memory bank width (AMD) 256 bytes
  23. Local memory syze per CU (AMD) 65536 (64KiB)
  24. Local memory banks (AMD) 32
  25. Preferred constant buffer size (AMD) 16384 (16KiB)
  26. Profiling timer offset since Epoch (AMD) 1649459515503492156ns (Sat Apr 9 01:11:55 2022)
  27. Thread trace supported (AMD) No
  28. Number of async queues (AMD) 2
  29. Max real-time compute queues (AMD) 0
  30. Max real-time compute units (AMD) 4174848
  31. Device Extensions cl_khr_fp64 cl_amd_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_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event
  32. Device Name Tahiti
  33. Device Version OpenCL 1.2 AMD-APP (3224.4)
  34. Driver Version 3224.4
  35. Device Type GPU
  36. Device Board Name (AMD) ATI FirePro V (FireGL V) Graphics Adapter
  37. Device Topology (AMD) PCI-E, 04:00.0
  38. SIMD per compute unit (AMD) 4
  39. SIMD width (AMD) 16
  40. SIMD instruction width (AMD) 1
  41. Graphics IP (AMD) 6.0
  42. Preferred work group size (AMD) 256
  43. Max work group size (AMD) 1024
  44. Wavefront width (AMD) 64
  45. Global free memory (AMD) <printDeviceInfo:78: get number of CL_DEVICE_GLOBAL_FREE_MEMORY_AMD : error -33>
  46. Global memory channels (AMD) 12
  47. Global memory banks per channel (AMD) 16
  48. Global memory bank width (AMD) 256 bytes
  49. Local memory syze per CU (AMD) 65536 (64KiB)
  50. Local memory banks (AMD) 32
  51. Preferred constant buffer size (AMD) 16384 (16KiB)
  52. Profiling timer offset since Epoch (AMD) 1649459515503492156ns (Sat Apr 9 01:11:55 2022)
  53. Thread trace supported (AMD) No
  54. Number of async queues (AMD) 2
  55. Max real-time compute queues (AMD) 0
  56. Max real-time compute units (AMD) 1702125412
  57. Device Extensions cl_khr_fp64 cl_amd_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_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event
  58. clCreateContext(NULL, ...) [other] Success [AMD]
  59.  
  60.  
  61. ## GCN2, Orca
  62.  
  63. Device Name AMD Radeon (TM) R9 390 Series (HAWAII, DRM 3.35.0, 5.4.0-44-generic, LLVM 10.0.0)
  64. Platform Name AMD Accelerated Parallel Processing
  65. Device Name Hawaii
  66. Device Version OpenCL 1.2 AMD-APP (3110.6)
  67. Driver Version 3110.6
  68. Device Type GPU
  69. Device Board Name (AMD) AMD Radeon (TM) R9 390 Series
  70. Device Topology (AMD) PCI-E, 01:00.0
  71. SIMD per compute unit (AMD) 4
  72. SIMD width (AMD) 16
  73. SIMD instruction width (AMD) 1
  74. Graphics IP (AMD) 7.2
  75. Preferred work group size (AMD) 256
  76. Max work group size (AMD) 1024
  77. Wavefront width (AMD) 64
  78. Global free memory (AMD) 6948212 (6.626GiB)
  79. Global memory channels (AMD) 16
  80. Global memory banks per channel (AMD) 16
  81. Global memory bank width (AMD) 256 bytes
  82. Local memory syze per CU (AMD) 65536 (64KiB)
  83. Local memory banks (AMD) 32
  84. Preferred constant buffer size (AMD) 16384 (16KiB)
  85. Profiling timer offset since Epoch (AMD) 1598554401773379990ns (Thu Aug 27 20:53:21 2020)
  86. Thread trace supported (AMD) Yes
  87. Number of async queues (AMD) 2
  88. Max real-time compute queues (AMD) 0
  89. Max real-time compute units (AMD) 0
  90. Device Extensions cl_khr_fp64 cl_amd_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_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event
  91.  
  92.  
  93. ## GCN4, PAL
  94.  
  95. Device Name Radeon RX 570 Series (POLARIS10, DRM 3.35.0, 5.4.0-42-generic, LLVM 10.0.0)
  96. Platform Name AMD Accelerated Parallel Processing
  97. Device Name Ellesmere
  98. Device Version OpenCL 1.2 AMD-APP (3110.6)
  99. Driver Version 3110.6
  100. Device Type GPU
  101. Device Board Name (AMD) Radeon RX 570 Series
  102. Device Topology (AMD) PCI-E, 41:00.0
  103. SIMD per compute unit (AMD) 4
  104. SIMD width (AMD) 16
  105. SIMD instruction width (AMD) 1
  106. Graphics IP (AMD) 8.0
  107. Preferred work group size (AMD) 256
  108. Max work group size (AMD) 1024
  109. Wavefront width (AMD) 64
  110. Global free memory (AMD) 3625616 (3.458GiB)
  111. Global memory channels (AMD) 8
  112. Global memory banks per channel (AMD) 16
  113. Global memory bank width (AMD) 256 bytes
  114. Local memory syze per CU (AMD) 65536 (64KiB)
  115. Local memory banks (AMD) 32
  116. Preferred constant buffer size (AMD) 16384 (16KiB)
  117. Profiling timer offset since Epoch (AMD) 1599064567445515778ns (Wed Sep 2 18:36:07 2020)
  118. Thread trace supported (AMD) Yes
  119. Number of async queues (AMD) 2
  120. Max real-time compute queues (AMD) 0
  121. Max real-time compute units (AMD) 0
  122. Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event
  123.  
  124.  
  125.  
  126. ## GCN5, PAL
  127.  
  128. Device Name AMD Radeon(TM) Vega 8 Graphics (RAVEN, DRM 3.41.0, 5.13.0-44-generic, LLVM 9.0.1)
  129. Platform Name AMD Accelerated Parallel Processing
  130. Device Name gfx902
  131. Device Version OpenCL 2.0 AMD-APP (3180.7)
  132. Driver Version 3180.7 (PAL,HSAIL)
  133. Device Type GPU
  134. Device Board Name (AMD) AMD Radeon(TM) Vega 8 Graphics
  135. Device Topology (AMD) PCI-E, 06:00.0
  136. SIMD per compute unit (AMD) 4
  137. SIMD width (AMD) 16
  138. SIMD instruction width (AMD) 1
  139. Graphics IP (AMD) 9.2
  140. Preferred work group size (AMD) 256
  141. Max work group size (AMD) 1024
  142. Wavefront width (AMD) 64
  143. Global free memory (AMD) 3600288 (3.434GiB)
  144. Global memory channels (AMD) 4
  145. Global memory banks per channel (AMD) 4
  146. Global memory bank width (AMD) 256 bytes
  147. Local memory syze per CU (AMD) 65536 (64KiB)
  148. Local memory banks (AMD) 32
  149. Preferred constant buffer size (AMD) 16384 (16KiB)
  150. Number of P2P devices (AMD) 0
  151. P2P devices (AMD) <printDeviceInfo:147: get number of CL_DEVICE_P2P_DEVICES_AMD : error -30>
  152. Profiling timer offset since Epoch (AMD) 1653402785834971031ns (Tue May 24 16:33:05 2022)
  153. Thread trace supported (AMD) Yes
  154. Number of async queues (AMD) 4
  155. Max real-time compute queues (AMD) 1
  156. Max real-time compute units (AMD) 0
  157. Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_sharing cl_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p
  158.  
  159.  
  160. Device Name AMD RAVEN (DRM 3.41.0, 5.13.0-39-generic, LLVM 9.0.1)
  161. Platform Name AMD Accelerated Parallel Processing
  162. Device Name gfx902
  163. Device Version OpenCL 2.0 AMD-APP (3180.7)
  164. Driver Version 3180.7 (PAL,HSAIL)
  165. Device Type GPU
  166. Device Board Name (AMD) Unknown AMD GPU
  167. Device Topology (AMD) PCI-E, 0a:00.0
  168. SIMD per compute unit (AMD) 4
  169. SIMD width (AMD) 16
  170. SIMD instruction width (AMD) 1
  171. Graphics IP (AMD) 9.2
  172. Preferred work group size (AMD) 256
  173. Max work group size (AMD) 1024
  174. Wavefront width (AMD) 64
  175. Global free memory (AMD) 3600288 (3.434GiB)
  176. Global memory channels (AMD) 4
  177. Global memory banks per channel (AMD) 4
  178. Global memory bank width (AMD) 256 bytes
  179. Local memory syze per CU (AMD) 65536 (64KiB)
  180. Local memory banks (AMD) 32
  181. Preferred constant buffer size (AMD) 16384 (16KiB)
  182. Number of P2P devices (AMD) 0
  183. P2P devices (AMD) <printDeviceInfo:147: get number of CL_DEVICE_P2P_DEVICES_AMD : error -30>
  184. Profiling timer offset since Epoch (AMD) 1649459515503492134ns (Sat Apr 9 01:11:55 2022)
  185. Thread trace supported (AMD) Yes
  186. Number of async queues (AMD) 4
  187. Max real-time compute queues (AMD) 1
  188. Max real-time compute units (AMD) 0
  189. Device Extensions cl_khr_fp64 cl_amd_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_khr_gl_sharing cl_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_copy_buffer_p2p
  190.  
  191.  
  192. ## GCN5, ROCm
  193.  
  194. Device Name AMD RAVEN (LLVM 13.0.1, DRM 3.42, 5.15.0-52-generic)
  195. Device Name gfx902:xnack-
  196. Device Version OpenCL 2.0
  197. Driver Version 3486.0 (HSA1.1,LC)
  198. Device Type GPU
  199. Device Board Name (AMD) AMD Ryzen 3 3200G with Radeon Vega Graphics
  200. Device PCI-e ID (AMD) 0x15d8
  201. Device Topology (AMD) PCI-E, 0000:09:00.0
  202. SIMD per compute unit (AMD) 4
  203. SIMD width (AMD) 16
  204. SIMD instruction width (AMD) 1
  205. Graphics IP (AMD) 9.0
  206. Preferred work group size (AMD) 256
  207. Max work group size (AMD) 1024
  208. Wavefront width (AMD) 64
  209. Global free memory (AMD) 15363024 (14.65GiB) 15363024 (14.65GiB)
  210. Global memory channels (AMD) 2
  211. Global memory banks per channel (AMD) 4
  212. Global memory bank width (AMD) 256 bytes
  213. Local memory size per CU (AMD) 65536 (64KiB)
  214. Local memory banks (AMD) 32
  215. Preferred constant buffer size (AMD) 16384 (16KiB)
  216. Number of P2P devices (AMD) 1
  217. P2P devices (AMD) [ 0x55926529be20 ]
  218. Profiling timer offset since Epoch (AMD) 0ns (Thu Jan 1 01:00:00 1970)
  219. Thread trace supported (AMD) No
  220. Number of async queues (AMD) 8
  221. Max real-time compute queues (AMD) 8
  222. Max real-time compute units (AMD) 11
  223. Device 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_khr_gl_sharing 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
  224.  
  225.  
  226. Device Name Radeon Instinct MI25 (vega10, LLVM 13.0.1, DRM 3.42, 5.15.0-52-generic)
  227. Device Name gfx900:xnack-
  228. Device Version OpenCL 2.0
  229. Driver Version 3486.0 (HSA1.1,LC)
  230. Device Type GPU
  231. Device Board Name (AMD) Radeon Instinct MI25
  232. Device PCI-e ID (AMD) 0x6860
  233. Device Topology (AMD) PCI-E, 0000:03:00.0
  234. SIMD per compute unit (AMD) 4
  235. SIMD width (AMD) 16
  236. SIMD instruction width (AMD) 1
  237. Graphics IP (AMD) 9.0
  238. Preferred work group size (AMD) 256
  239. Max work group size (AMD) 1024
  240. Wavefront width (AMD) 64
  241. Global free memory (AMD) 16760832 (15.98GiB) 16760832 (15.98GiB)
  242. Global memory channels (AMD) 64
  243. Global memory banks per channel (AMD) 4
  244. Global memory bank width (AMD) 256 bytes
  245. Local memory size per CU (AMD) 65536 (64KiB)
  246. Local memory banks (AMD) 32
  247. Preferred constant buffer size (AMD) 16384 (16KiB)
  248. Number of P2P devices (AMD) 0
  249. Profiling timer offset since Epoch (AMD) 0ns (Thu Jan 1 01:00:00 1970)
  250. Thread trace supported (AMD) No
  251. Number of async queues (AMD) 8
  252. Max real-time compute queues (AMD) 8
  253. Max real-time compute units (AMD) 64
  254. Device 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_khr_gl_sharing 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
  255. 4
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement