API tools faq
paste
Vladar

vk_pro-progl-clinfo

Vladar
Nov 28th, 2024
32
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 42.25 KB | None | 0 0
raw download clone embed print report
  1. Number of platforms 4
  2. Platform Name AMD Accelerated Parallel Processing
  3. Platform Vendor Advanced Micro Devices, Inc.
  4. Platform Version OpenCL 2.1 AMD-APP (3590.0)
  5. Platform Profile FULL_PROFILE
  6. Platform Extensions cl_khr_icd cl_amd_event_callback
  7. Platform Extensions function suffix AMD
  8. Platform Host timer resolution 1ns
  9.  
  10. Platform Name Intel(R) OpenCL
  11. Platform Vendor Intel(R) Corporation
  12. Platform Version OpenCL 3.0 LINUX
  13. Platform Profile FULL_PROFILE
  14. Platform Extensions cl_khr_spirv_linkonce_odr cl_khr_fp64 cl_khr_fp16 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_3d_image_writes cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_icd cl_khr_il_program cl_intel_unified_shared_memory cl_intel_devicelib_assert cl_khr_subgroup_shuffle cl_khr_subgroup_shuffle_relative cl_khr_subgroup_extended_types cl_khr_subgroup_non_uniform_arithmetic cl_intel_subgroups cl_intel_subgroups_char cl_intel_subgroups_short cl_intel_subgroups_long cl_intel_required_subgroup_size cl_intel_spirv_subgroups cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_intel_device_attribute_query cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_image2d_from_buffer cl_intel_concurrent_dispatch
  15. Platform Extensions with Version cl_khr_spirv_linkonce_odr 0x400000 (1.0.0)
  16. cl_khr_fp64 0x400000 (1.0.0)
  17. cl_khr_fp16 0x400000 (1.0.0)
  18. cl_khr_global_int32_base_atomics 0x400000 (1.0.0)
  19. cl_khr_global_int32_extended_atomics 0x400000 (1.0.0)
  20. cl_khr_local_int32_base_atomics 0x400000 (1.0.0)
  21. cl_khr_local_int32_extended_atomics 0x400000 (1.0.0)
  22. cl_khr_3d_image_writes 0x400000 (1.0.0)
  23. cl_khr_byte_addressable_store 0x400000 (1.0.0)
  24. cl_khr_depth_images 0x400000 (1.0.0)
  25. cl_khr_icd 0x400000 (1.0.0)
  26. cl_khr_il_program 0x400000 (1.0.0)
  27. cl_intel_unified_shared_memory 0x400000 (1.0.0)
  28. cl_intel_devicelib_assert 0x400000 (1.0.0)
  29. cl_khr_subgroup_shuffle 0x400000 (1.0.0)
  30. cl_khr_subgroup_shuffle_relative 0x400000 (1.0.0)
  31. cl_khr_subgroup_extended_types 0x400000 (1.0.0)
  32. cl_khr_subgroup_non_uniform_arithmetic 0x400000 (1.0.0)
  33. cl_intel_subgroups 0x400000 (1.0.0)
  34. cl_intel_subgroups_char 0x400000 (1.0.0)
  35. cl_intel_subgroups_short 0x400000 (1.0.0)
  36. cl_intel_subgroups_long 0x400000 (1.0.0)
  37. cl_intel_required_subgroup_size 0x400000 (1.0.0)
  38. cl_intel_spirv_subgroups 0x400000 (1.0.0)
  39. cl_khr_int64_base_atomics 0x400000 (1.0.0)
  40. cl_khr_int64_extended_atomics 0x400000 (1.0.0)
  41. cl_intel_device_attribute_query 0x400000 (1.0.0)
  42. cl_intel_exec_by_local_thread 0x400000 (1.0.0)
  43. cl_intel_vec_len_hint 0x400000 (1.0.0)
  44. cl_intel_device_partition_by_names 0x400000 (1.0.0)
  45. cl_khr_spir 0x400000 (1.0.0)
  46. cl_khr_image2d_from_buffer 0x400000 (1.0.0)
  47. cl_intel_concurrent_dispatch 0x400000 (1.0.0)
  48. Platform Numeric Version 0xc00000 (3.0.0)
  49. Platform Extensions function suffix INTEL
  50. Platform Host timer resolution 1ns
  51.  
  52. Platform Name rusticl
  53. Platform Vendor Mesa/X.org
  54. Platform Version OpenCL 3.0
  55. Platform Profile FULL_PROFILE
  56. Platform Extensions cl_khr_byte_addressable_store cl_khr_create_command_queue cl_khr_expect_assume cl_khr_extended_versioning cl_khr_icd cl_khr_il_program cl_khr_spirv_no_integer_wrap_decoration cl_khr_suggested_local_work_size
  57. Platform Extensions with Version cl_khr_byte_addressable_store 0x400000 (1.0.0)
  58. cl_khr_create_command_queue 0x400000 (1.0.0)
  59. cl_khr_expect_assume 0x400000 (1.0.0)
  60. cl_khr_extended_versioning 0x400000 (1.0.0)
  61. cl_khr_icd 0x400000 (1.0.0)
  62. cl_khr_il_program 0x400000 (1.0.0)
  63. cl_khr_spirv_no_integer_wrap_decoration 0x400000 (1.0.0)
  64. cl_khr_suggested_local_work_size 0x400000 (1.0.0)
  65. Platform Numeric Version 0xc00000 (3.0.0)
  66. Platform Extensions function suffix MESA
  67. Platform Host timer resolution 1ns
  68.  
  69. Platform Name AMD Accelerated Parallel Processing
  70. Platform Vendor Advanced Micro Devices, Inc.
  71. Platform Version OpenCL 2.1 AMD-APP (3380.4)
  72. Platform Profile FULL_PROFILE
  73. Platform Extensions cl_khr_icd cl_amd_event_callback cl_amd_offline_devices
  74. Platform Extensions function suffix AMD
  75. Platform Host timer resolution 1ns
  76.  
  77. Platform Name AMD Accelerated Parallel Processing
  78. Number of devices 1
  79. Device Name gfx803
  80. Device Vendor Advanced Micro Devices, Inc.
  81. Device Vendor ID 0x1002
  82. Device Version OpenCL 1.2
  83. Driver Version 3590.0 (HSA1.1,LC)
  84. Device OpenCL C Version OpenCL C 2.0
  85. Device Type GPU
  86. Device Board Name (AMD) AMD Radeon RX 580 Series
  87. Device PCI-e ID (AMD) 0x67df
  88. Device Topology (AMD) PCI-E, 0000:01:00.0
  89. Device Profile FULL_PROFILE
  90. Device Available Yes
  91. Compiler Available Yes
  92. Linker Available Yes
  93. Max compute units 36
  94. SIMD per compute unit (AMD) 4
  95. SIMD width (AMD) 16
  96. SIMD instruction width (AMD) 1
  97. Max clock frequency 1340MHz
  98. Graphics IP (AMD) 8.0
  99. Device Partition (core)
  100. Max number of sub-devices 36
  101. Supported partition types None
  102. Supported affinity domains (n/a)
  103. Max work item dimensions 3
  104. Max work item sizes 1024x1024x1024
  105. Max work group size 256
  106. Preferred work group size (AMD) 256
  107. Max work group size (AMD) 1024
  108. Preferred work group size multiple (kernel) <getWGsizes:1980: create kernel : error -6>
  109. Wavefront width (AMD) 64
  110. Preferred / native vector sizes
  111. char 4 / 4
  112. short 2 / 2
  113. int 1 / 1
  114. long 1 / 1
  115. half 1 / 1 (cl_khr_fp16)
  116. float 1 / 1
  117. double 1 / 1 (cl_khr_fp64)
  118. Half-precision Floating-point support (cl_khr_fp16)
  119. Denormals No
  120. Infinity and NANs Yes
  121. Round to nearest Yes
  122. Round to zero Yes
  123. Round to infinity Yes
  124. IEEE754-2008 fused multiply-add Yes
  125. Support is emulated in software No
  126. Single-precision Floating-point support (core)
  127. Denormals No
  128. Infinity and NANs Yes
  129. Round to nearest Yes
  130. Round to zero Yes
  131. Round to infinity Yes
  132. IEEE754-2008 fused multiply-add Yes
  133. Support is emulated in software No
  134. Correctly-rounded divide and sqrt operations Yes
  135. Double-precision Floating-point support (cl_khr_fp64)
  136. Denormals Yes
  137. Infinity and NANs Yes
  138. Round to nearest Yes
  139. Round to zero Yes
  140. Round to infinity Yes
  141. IEEE754-2008 fused multiply-add Yes
  142. Support is emulated in software No
  143. Address bits 64, Little-Endian
  144. Global memory size 8589934592 (8GiB)
  145. Global free memory (AMD) 8364032 (7.977GiB) 8364032 (7.977GiB)
  146. Global memory channels (AMD) 8
  147. Global memory banks per channel (AMD) 4
  148. Global memory bank width (AMD) 256 bytes
  149. Error Correction support No
  150. Max memory allocation 7301444400 (6.8GiB)
  151. Unified memory for Host and Device No
  152. Minimum alignment for any data type 128 bytes
  153. Alignment of base address 1024 bits (128 bytes)
  154. Global Memory cache type Read/Write
  155. Global Memory cache size 16384 (16KiB)
  156. Global Memory cache line size 64 bytes
  157. Image support Yes
  158. Max number of samplers per kernel 16
  159. Max size for 1D images from buffer 134217728 pixels
  160. Max 1D or 2D image array size 8192 images
  161. Base address alignment for 2D image buffers 256 bytes
  162. Pitch alignment for 2D image buffers 256 pixels
  163. Max 2D image size 16384x16384 pixels
  164. Max 3D image size 16384x16384x8192 pixels
  165. Max number of read image args 128
  166. Max number of write image args 8
  167. Local memory type Local
  168. Local memory size 65536 (64KiB)
  169. Local memory size per CU (AMD) 65536 (64KiB)
  170. Local memory banks (AMD) 32
  171. Max number of constant args 8
  172. Max constant buffer size 7301444400 (6.8GiB)
  173. Preferred constant buffer size (AMD) 16384 (16KiB)
  174. Max size of kernel argument 1024
  175. Queue properties
  176. Out-of-order execution No
  177. Profiling Yes
  178. Prefer user sync for interop Yes
  179. Number of P2P devices (AMD) 0
  180. Profiling timer resolution 1ns
  181. Profiling timer offset since Epoch (AMD) 0ns (Thu Jan 1 03:00:00 1970)
  182. Execution capabilities
  183. Run OpenCL kernels Yes
  184. Run native kernels No
  185. Thread trace supported (AMD) No
  186. Number of async queues (AMD) 8
  187. Max real-time compute queues (AMD) 8
  188. Max real-time compute units (AMD) 36
  189. printf() buffer size 4194304 (4MiB)
  190. Built-in kernels (n/a)
  191. 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
  192.  
  193. Platform Name Intel(R) OpenCL
  194. Number of devices 1
  195. Device Name Intel(R) Core(TM) i5-7400 CPU @ 3.00GHz
  196. Device Vendor Intel(R) Corporation
  197. Device Vendor ID 0x8086
  198. Device Version OpenCL 3.0 (Build 0)
  199. Device Numeric Version 0xc00000 (3.0.0)
  200. Driver Version 2024.17.3.0.08_160000
  201. Device OpenCL C Version OpenCL C 3.0
  202. Device OpenCL C all versions OpenCL C 0xc00000 (3.0.0)
  203. OpenCL C 0x800000 (2.0.0)
  204. OpenCL C 0x402000 (1.2.0)
  205. OpenCL C 0x401000 (1.1.0)
  206. OpenCL C 0x400000 (1.0.0)
  207. Device OpenCL C features __opencl_c_3d_image_writes 0xc00000 (3.0.0)
  208. __opencl_c_atomic_order_acq_rel 0xc00000 (3.0.0)
  209. __opencl_c_atomic_order_seq_cst 0xc00000 (3.0.0)
  210. __opencl_c_atomic_scope_device 0xc00000 (3.0.0)
  211. __opencl_c_atomic_scope_all_devices 0xc00000 (3.0.0)
  212. __opencl_c_device_enqueue 0xc00000 (3.0.0)
  213. __opencl_c_generic_address_space 0xc00000 (3.0.0)
  214. __opencl_c_fp64 0xc00000 (3.0.0)
  215. __opencl_c_images 0xc00000 (3.0.0)
  216. __opencl_c_int64 0xc00000 (3.0.0)
  217. __opencl_c_pipes 0xc00000 (3.0.0)
  218. __opencl_c_program_scope_global_variables 0xc00000 (3.0.0)
  219. __opencl_c_read_write_images 0xc00000 (3.0.0)
  220. __opencl_c_subgroups 0xc00000 (3.0.0)
  221. __opencl_c_work_group_collective_functions 0xc00000 (3.0.0)
  222. Latest conformance test passed v2023-10-10-00
  223. Device Type CPU
  224. Device Profile FULL_PROFILE
  225. Device Available Yes
  226. Compiler Available Yes
  227. Linker Available Yes
  228. Max compute units 4
  229. Max clock frequency 3000MHz
  230. Device Partition (core)
  231. Max number of sub-devices 4
  232. Supported partition types by counts, equally, by names (Intel)
  233. Supported affinity domains (n/a)
  234. Max work item dimensions 3
  235. Max work item sizes 8192x8192x8192
  236. Max work group size 8192
  237. Preferred work group size multiple (device) 128
  238. Preferred work group size multiple (kernel) 128
  239. Max sub-groups per work group 2048
  240. Sub-group sizes (Intel) 4, 8, 16, 32, 64
  241. Preferred / native vector sizes
  242. char 1 / 32
  243. short 1 / 16
  244. int 1 / 8
  245. long 1 / 4
  246. half 0 / 0 (cl_khr_fp16)
  247. float 1 / 8
  248. double 1 / 4 (cl_khr_fp64)
  249. Half-precision Floating-point support (cl_khr_fp16)
  250. Denormals No
  251. Infinity and NANs Yes
  252. Round to nearest Yes
  253. Round to zero No
  254. Round to infinity No
  255. IEEE754-2008 fused multiply-add No
  256. Support is emulated in software No
  257. Single-precision Floating-point support (core)
  258. Denormals Yes
  259. Infinity and NANs Yes
  260. Round to nearest Yes
  261. Round to zero No
  262. Round to infinity No
  263. IEEE754-2008 fused multiply-add No
  264. Support is emulated in software No
  265. Correctly-rounded divide and sqrt operations No
  266. Double-precision Floating-point support (cl_khr_fp64)
  267. Denormals Yes
  268. Infinity and NANs Yes
  269. Round to nearest Yes
  270. Round to zero Yes
  271. Round to infinity Yes
  272. IEEE754-2008 fused multiply-add Yes
  273. Support is emulated in software No
  274. Address bits 64, Little-Endian
  275. Global memory size 33591115776 (31.28GiB)
  276. Error Correction support No
  277. Max memory allocation 16795557888 (15.64GiB)
  278. Unified memory for Host and Device Yes
  279. Shared Virtual Memory (SVM) capabilities (core)
  280. Coarse-grained buffer sharing Yes
  281. Fine-grained buffer sharing Yes
  282. Fine-grained system sharing Yes
  283. Atomics Yes
  284. Unified Shared Memory (USM) (cl_intel_unified_shared_memory)
  285. Host USM capabilities (Intel) USM access, USM atomic access, USM concurrent access, USM concurrent atomic access
  286. Device USM capabilities (Intel) USM access, USM atomic access, USM concurrent access, USM concurrent atomic access
  287. Single-Device USM caps (Intel) USM access, USM atomic access, USM concurrent access, USM concurrent atomic access
  288. Cross-Device USM caps (Intel) USM access, USM atomic access, USM concurrent access, USM concurrent atomic access
  289. Shared System USM caps (Intel) USM access, USM atomic access, USM concurrent access, USM concurrent atomic access
  290. Minimum alignment for any data type 128 bytes
  291. Alignment of base address 1024 bits (128 bytes)
  292. Preferred alignment for atomics
  293. SVM 64 bytes
  294. Global 64 bytes
  295. Local 0 bytes
  296. Atomic memory capabilities relaxed, acquire/release, sequentially-consistent, work-group scope, device scope, all-devices scope
  297. Atomic fence capabilities relaxed, acquire/release, sequentially-consistent, work-item scope, work-group scope, device scope, all-devices scope
  298. Max size for global variable 65536 (64KiB)
  299. Preferred total size of global vars 65536 (64KiB)
  300. Global Memory cache type Read/Write
  301. Global Memory cache size 262144 (256KiB)
  302. Global Memory cache line size 64 bytes
  303. Image support Yes
  304. Max number of samplers per kernel 480
  305. Max size for 1D images from buffer 1049722368 pixels
  306. Max 1D or 2D image array size 2048 images
  307. Base address alignment for 2D image buffers 64 bytes
  308. Pitch alignment for 2D image buffers 64 pixels
  309. Max 2D image size 16384x16384 pixels
  310. Max 3D image size 2048x2048x2048 pixels
  311. Max number of read image args 480
  312. Max number of write image args 480
  313. Max number of read/write image args 480
  314. Pipe support Yes
  315. Max number of pipe args 16
  316. Max active pipe reservations 65535
  317. Max pipe packet size 1024
  318. Local memory type Global
  319. Local memory size 32768 (32KiB)
  320. Max number of constant args 480
  321. Max constant buffer size 131072 (128KiB)
  322. Generic address space support Yes
  323. Max size of kernel argument 3840 (3.75KiB)
  324. Queue properties (on host)
  325. Out-of-order execution Yes
  326. Profiling Yes
  327. Local thread execution (Intel) Yes
  328. Device enqueue capabilities supported, replaceable default queue
  329. Queue properties (on device)
  330. Out-of-order execution Yes
  331. Profiling Yes
  332. Preferred size 4294967295 (4GiB)
  333. Max size 4294967295 (4GiB)
  334. Max queues on device 4294967295
  335. Max events on device 4294967295
  336. Prefer user sync for interop No
  337. Profiling timer resolution 1ns
  338. Execution capabilities
  339. Run OpenCL kernels Yes
  340. Run native kernels Yes
  341. Non-uniform work-groups Yes
  342. Work-group collective functions Yes
  343. Sub-group independent forward progress No
  344. IL version SPIR-V_1.0
  345. ILs with version SPIR-V 0x400000 (1.0.0)
  346. SPIR versions 1.2
  347. printf() buffer size 1048576 (1024KiB)
  348. Built-in kernels (n/a)
  349. Built-in kernels with version (n/a)
  350. Device Extensions cl_khr_spirv_linkonce_odr cl_khr_fp64 cl_khr_fp16 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_3d_image_writes cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_icd cl_khr_il_program cl_intel_unified_shared_memory cl_intel_devicelib_assert cl_khr_subgroup_shuffle cl_khr_subgroup_shuffle_relative cl_khr_subgroup_extended_types cl_khr_subgroup_non_uniform_arithmetic cl_intel_subgroups cl_intel_subgroups_char cl_intel_subgroups_short cl_intel_subgroups_long cl_intel_required_subgroup_size cl_intel_spirv_subgroups cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_intel_device_attribute_query cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_image2d_from_buffer cl_intel_concurrent_dispatch
  351. Device Extensions with Version cl_khr_spirv_linkonce_odr 0x400000 (1.0.0)
  352. cl_khr_fp64 0x400000 (1.0.0)
  353. cl_khr_fp16 0x400000 (1.0.0)
  354. cl_khr_global_int32_base_atomics 0x400000 (1.0.0)
  355. cl_khr_global_int32_extended_atomics 0x400000 (1.0.0)
  356. cl_khr_local_int32_base_atomics 0x400000 (1.0.0)
  357. cl_khr_local_int32_extended_atomics 0x400000 (1.0.0)
  358. cl_khr_3d_image_writes 0x400000 (1.0.0)
  359. cl_khr_byte_addressable_store 0x400000 (1.0.0)
  360. cl_khr_depth_images 0x400000 (1.0.0)
  361. cl_khr_icd 0x400000 (1.0.0)
  362. cl_khr_il_program 0x400000 (1.0.0)
  363. cl_intel_unified_shared_memory 0x400000 (1.0.0)
  364. cl_intel_devicelib_assert 0x400000 (1.0.0)
  365. cl_khr_subgroup_shuffle 0x400000 (1.0.0)
  366. cl_khr_subgroup_shuffle_relative 0x400000 (1.0.0)
  367. cl_khr_subgroup_extended_types 0x400000 (1.0.0)
  368. cl_khr_subgroup_non_uniform_arithmetic 0x400000 (1.0.0)
  369. cl_intel_subgroups 0x400000 (1.0.0)
  370. cl_intel_subgroups_char 0x400000 (1.0.0)
  371. cl_intel_subgroups_short 0x400000 (1.0.0)
  372. cl_intel_subgroups_long 0x400000 (1.0.0)
  373. cl_intel_required_subgroup_size 0x400000 (1.0.0)
  374. cl_intel_spirv_subgroups 0x400000 (1.0.0)
  375. cl_khr_int64_base_atomics 0x400000 (1.0.0)
  376. cl_khr_int64_extended_atomics 0x400000 (1.0.0)
  377. cl_intel_device_attribute_query 0x400000 (1.0.0)
  378. cl_intel_exec_by_local_thread 0x400000 (1.0.0)
  379. cl_intel_vec_len_hint 0x400000 (1.0.0)
  380. cl_intel_device_partition_by_names 0x400000 (1.0.0)
  381. cl_khr_spir 0x400000 (1.0.0)
  382. cl_khr_image2d_from_buffer 0x400000 (1.0.0)
  383. cl_intel_concurrent_dispatch 0x400000 (1.0.0)
  384.  
  385.  
  386. Platform Name rusticl
  387. Number of devices 0
  388.  
  389. Platform Name AMD Accelerated Parallel Processing
  390. Number of devices 1
  391. Device Name Ellesmere
  392. Device Vendor Advanced Micro Devices, Inc.
  393. Device Vendor ID 0x1002
  394. Device Version OpenCL 2.0 AMD-APP (3380.4)
  395. Driver Version 3380.4 (PAL,HSAIL)
  396. Device OpenCL C Version OpenCL C 2.0
  397. Device Type GPU
  398. Device Board Name (AMD) AMD Radeon RX 580 Series
  399. Device PCI-e ID (AMD) 0x67df
  400. Device Topology (AMD) PCI-E, 0000:01:00.0
  401. Device Profile FULL_PROFILE
  402. Device Available Yes
  403. Compiler Available Yes
  404. Linker Available Yes
  405. Max compute units 36
  406. SIMD per compute unit (AMD) 4
  407. SIMD width (AMD) 16
  408. SIMD instruction width (AMD) 1
  409. Max clock frequency 1340MHz
  410. Graphics IP (AMD) 8.0
  411. Device Partition (core)
  412. Max number of sub-devices 36
  413. Supported partition types None
  414. Supported affinity domains (n/a)
  415. Max work item dimensions 3
  416. Max work item sizes 1024x1024x1024
  417. Max work group size 256
  418. Preferred work group size (AMD) 256
  419. Max work group size (AMD) 1024
  420. Preferred work group size multiple (kernel) 64
  421. Wavefront width (AMD) 64
  422. Preferred / native vector sizes
  423. char 4 / 4
  424. short 2 / 2
  425. int 1 / 1
  426. long 1 / 1
  427. half 1 / 1 (cl_khr_fp16)
  428. float 1 / 1
  429. double 1 / 1 (cl_khr_fp64)
  430. Half-precision Floating-point support (cl_khr_fp16)
  431. Denormals No
  432. Infinity and NANs No
  433. Round to nearest No
  434. Round to zero No
  435. Round to infinity No
  436. IEEE754-2008 fused multiply-add No
  437. Support is emulated in software No
  438. Single-precision Floating-point support (core)
  439. Denormals No
  440. Infinity and NANs Yes
  441. Round to nearest Yes
  442. Round to zero Yes
  443. Round to infinity Yes
  444. IEEE754-2008 fused multiply-add Yes
  445. Support is emulated in software No
  446. Correctly-rounded divide and sqrt operations Yes
  447. Double-precision Floating-point support (cl_khr_fp64)
  448. Denormals Yes
  449. Infinity and NANs Yes
  450. Round to nearest Yes
  451. Round to zero Yes
  452. Round to infinity Yes
  453. IEEE754-2008 fused multiply-add Yes
  454. Support is emulated in software No
  455. Address bits 64, Little-Endian
  456. Global memory size 8589934592 (8GiB)
  457. Global free memory (AMD) 8323072 (7.938GiB) 8060928 (7.688GiB)
  458. Global memory channels (AMD) 8
  459. Global memory banks per channel (AMD) 4
  460. Global memory bank width (AMD) 256 bytes
  461. Error Correction support No
  462. Max memory allocation 7073274265 (6.587GiB)
  463. Unified memory for Host and Device No
  464. Shared Virtual Memory (SVM) capabilities (core)
  465. Coarse-grained buffer sharing Yes
  466. Fine-grained buffer sharing Yes
  467. Fine-grained system sharing No
  468. Atomics No
  469. Minimum alignment for any data type 128 bytes
  470. Alignment of base address 2048 bits (256 bytes)
  471. Preferred alignment for atomics
  472. SVM 0 bytes
  473. Global 0 bytes
  474. Local 0 bytes
  475. Max size for global variable 6365946624 (5.929GiB)
  476. Preferred total size of global vars 8589934592 (8GiB)
  477. Global Memory cache type Read/Write
  478. Global Memory cache size 16384 (16KiB)
  479. Global Memory cache line size 64 bytes
  480. Image support Yes
  481. Max number of samplers per kernel 16
  482. Max size for 1D images from buffer 442079641 pixels
  483. Max 1D or 2D image array size 2048 images
  484. Base address alignment for 2D image buffers 256 bytes
  485. Pitch alignment for 2D image buffers 256 pixels
  486. Max 2D image size 16384x16384 pixels
  487. Max 3D image size 2048x2048x2048 pixels
  488. Max number of read image args 128
  489. Max number of write image args 64
  490. Max number of read/write image args 64
  491. Max number of pipe args 16
  492. Max active pipe reservations 16
  493. Max pipe packet size 2778306969 (2.587GiB)
  494. Local memory type Local
  495. Local memory size 65536 (64KiB)
  496. Local memory size per CU (AMD) 65536 (64KiB)
  497. Local memory banks (AMD) 32
  498. Max number of constant args 8
  499. Max constant buffer size 7073274265 (6.587GiB)
  500. Preferred constant buffer size (AMD) 16384 (16KiB)
  501. Max size of kernel argument 1024
  502. Queue properties (on host)
  503. Out-of-order execution No
  504. Profiling Yes
  505. Queue properties (on device)
  506. Out-of-order execution Yes
  507. Profiling Yes
  508. Preferred size 262144 (256KiB)
  509. Max size 8388608 (8MiB)
  510. Max queues on device 1
  511. Max events on device 1024
  512. Prefer user sync for interop Yes
  513. Number of P2P devices (AMD) 0
  514. Profiling timer resolution 1ns
  515. Profiling timer offset since Epoch (AMD) 1732785967490361152ns (Thu Nov 28 12:26:07 2024)
  516. Execution capabilities
  517. Run OpenCL kernels Yes
  518. Run native kernels No
  519. Thread trace supported (AMD) Yes
  520. Number of async queues (AMD) 4
  521. Max real-time compute queues (AMD) 1
  522. Max real-time compute units (AMD) 0
  523. printf() buffer size 4194304 (4MiB)
  524. Built-in kernels (n/a)
  525. 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
  526.  
  527.  
  528. NULL platform behavior
  529. clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
  530. clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) No platform
  531. clCreateContext(NULL, ...) [default] No platform
  532. clCreateContext(NULL, ...) [other] Success [AMD]
  533. clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
  534. Platform Name AMD Accelerated Parallel Processing
  535. Device Name gfx803
  536. clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
  537. clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
  538. Platform Name AMD Accelerated Parallel Processing
  539. Device Name gfx803
  540. clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
  541. clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform
  542. clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1)
  543. Platform Name AMD Accelerated Parallel Processing
  544. Device Name gfx803
  545.  
  546. ICD loader properties
  547. ICD loader Name Khronos OpenCL ICD Loader
  548. ICD loader Vendor Khronos Group
  549. ICD loader Version 3.0.5
  550. ICD loader Profile OpenCL 3.0
  551.  
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!