Coverage for cuda/core/_utils/driver_cu_result_explanations

2# SPDX-License-Identifier: Apache-2.0

4# CUDA Toolkit v13.1.1

5_FALLBACK_EXPLANATIONS = {

6 0: (

7 "The API call returned with no errors. In the case of query calls, this"

8 " also means that the operation being queried is complete (see"

9 " ::cuEventQuery() and ::cuStreamQuery())."

10 ),

11 1: (

12 "This indicates that one or more of the parameters passed to the API call"

13 " is not within an acceptable range of values."

14 ),

15 2: (

16 "The API call failed because it was unable to allocate enough memory or"

17 " other resources to perform the requested operation."

18 ),

19 3: (

20 "This indicates that the CUDA driver has not been initialized with"

21 " ::cuInit() or that initialization has failed."

22 ),

23 4: "This indicates that the CUDA driver is in the process of shutting down.",

24 5: (

25 "This indicates profiler is not initialized for this run. This can"

26 " happen when the application is running with external profiling tools"

27 " like visual profiler."

28 ),

29 6: (

30 "This error return is deprecated as of CUDA 5.0. It is no longer an error"

31 " to attempt to enable/disable the profiling via ::cuProfilerStart or"

32 " ::cuProfilerStop without initialization."

33 ),

34 7: (

35 "This error return is deprecated as of CUDA 5.0. It is no longer an error"

36 " to call cuProfilerStart() when profiling is already enabled."

37 ),

38 8: (

39 "This error return is deprecated as of CUDA 5.0. It is no longer an error"

40 " to call cuProfilerStop() when profiling is already disabled."

41 ),

42 34: (

43 "This indicates that the CUDA driver that the application has loaded is a"

44 " stub library. Applications that run with the stub rather than a real"

45 " driver loaded will result in CUDA API returning this error."

46 ),

47 36: (

48 "This indicates that the API call requires a newer CUDA driver than the one"

49 " currently installed. Users should install an updated NVIDIA CUDA driver"

50 " to allow the API call to succeed."

51 ),

52 46: (

53 "This indicates that requested CUDA device is unavailable at the current"

54 " time. Devices are often unavailable due to use of"

55 " ::CU_COMPUTEMODE_EXCLUSIVE_PROCESS or ::CU_COMPUTEMODE_PROHIBITED."

56 ),

57 100: ("This indicates that no CUDA-capable devices were detected by the installed CUDA driver."),

58 101: (

59 "This indicates that the device ordinal supplied by the user does not"

60 " correspond to a valid CUDA device or that the action requested is"

61 " invalid for the specified device."

62 ),

63 102: "This error indicates that the Grid license is not applied.",

64 200: ("This indicates that the device kernel image is invalid. This can also indicate an invalid CUDA module."),

65 201: (

66 "This most frequently indicates that there is no context bound to the"

67 " current thread. This can also be returned if the context passed to an"

68 " API call is not a valid handle (such as a context that has had"

69 " ::cuCtxDestroy() invoked on it). This can also be returned if a user"

70 " mixes different API versions (i.e. 3010 context with 3020 API calls)."

71 " See ::cuCtxGetApiVersion() for more details."

72 " This can also be returned if the green context passed to an API call"

73 " was not converted to a ::CUcontext using ::cuCtxFromGreenCtx API."

74 ),

75 202: (

76 "This indicated that the context being supplied as a parameter to the"

77 " API call was already the active context."

78 " This error return is deprecated as of CUDA 3.2. It is no longer an"

79 " error to attempt to push the active context via ::cuCtxPushCurrent()."

80 ),

81 205: "This indicates that a map or register operation has failed.",

82 206: "This indicates that an unmap or unregister operation has failed.",

83 207: ("This indicates that the specified array is currently mapped and thus cannot be destroyed."),

84 208: "This indicates that the resource is already mapped.",

85 209: (

86 "This indicates that there is no kernel image available that is suitable"

87 " for the device. This can occur when a user specifies code generation"

88 " options for a particular CUDA source file that do not include the"

89 " corresponding device configuration."

90 ),

91 210: "This indicates that a resource has already been acquired.",

92 211: "This indicates that a resource is not mapped.",

93 212: ("This indicates that a mapped resource is not available for access as an array."),

94 213: ("This indicates that a mapped resource is not available for access as a pointer."),

95 214: ("This indicates that an uncorrectable ECC error was detected during execution."),

96 215: ("This indicates that the ::CUlimit passed to the API call is not supported by the active device."),

97 216: (

98 "This indicates that the ::CUcontext passed to the API call can"

99 " only be bound to a single CPU thread at a time but is already"

100 " bound to a CPU thread."

101 ),

102 217: ("This indicates that peer access is not supported across the given devices."),

103 218: "This indicates that a PTX JIT compilation failed.",

104 219: "This indicates an error with OpenGL or DirectX context.",

105 220: ("This indicates that an uncorrectable NVLink error was detected during the execution."),

106 221: "This indicates that the PTX JIT compiler library was not found.",

107 222: "This indicates that the provided PTX was compiled with an unsupported toolchain.",

108 223: "This indicates that the PTX JIT compilation was disabled.",

109 224: ("This indicates that the ::CUexecAffinityType passed to the API call is not supported by the active device."),

110 225: (

111 "This indicates that the code to be compiled by the PTX JIT contains unsupported call to cudaDeviceSynchronize."

112 ),

113 226: (

114 "This indicates that an exception occurred on the device that is now"

115 " contained by the GPU's error containment capability. Common causes are -"

116 " a. Certain types of invalid accesses of peer GPU memory over nvlink"

117 " b. Certain classes of hardware errors"

118 " This leaves the process in an inconsistent state and any further CUDA"

119 " work will return the same error. To continue using CUDA, the process must"

120 " be terminated and relaunched."

121 ),

122 300: (

123 "This indicates that the device kernel source is invalid. This includes"

124 " compilation/linker errors encountered in device code or user error."

125 ),

126 301: "This indicates that the file specified was not found.",

127 302: "This indicates that a link to a shared object failed to resolve.",

128 303: "This indicates that initialization of a shared object failed.",

129 304: "This indicates that an OS call failed.",

130 400: (

131 "This indicates that a resource handle passed to the API call was not"

132 " valid. Resource handles are opaque types like ::CUstream and ::CUevent."

133 ),

134 401: (

135 "This indicates that a resource required by the API call is not in a"

136 " valid state to perform the requested operation."

137 ),

138 402: (

139 "This indicates an attempt was made to introspect an object in a way that"

140 " would discard semantically important information. This is either due to"

141 " the object using funtionality newer than the API version used to"

142 " introspect it or omission of optional return arguments."

143 ),

144 500: (

145 "This indicates that a named symbol was not found. Examples of symbols"

146 " are global/constant variable names, driver function names, texture names,"

147 " and surface names."

148 ),

149 600: (

150 "This indicates that asynchronous operations issued previously have not"

151 " completed yet. This result is not actually an error, but must be indicated"

152 " differently than ::CUDA_SUCCESS (which indicates completion). Calls that"

153 " may return this value include ::cuEventQuery() and ::cuStreamQuery()."

154 ),

155 700: (

156 "While executing a kernel, the device encountered a"

157 " load or store instruction on an invalid memory address."

158 " This leaves the process in an inconsistent state and any further CUDA work"

159 " will return the same error. To continue using CUDA, the process must be terminated"

160 " and relaunched."

161 ),

162 701: (

163 "This indicates that a launch did not occur because it did not have"

164 " appropriate resources. This error usually indicates that the user has"

165 " attempted to pass too many arguments to the device kernel, or the"

166 " kernel launch specifies too many threads for the kernel's register"

167 " count. Passing arguments of the wrong size (i.e. a 64-bit pointer"

168 " when a 32-bit int is expected) is equivalent to passing too many"

169 " arguments and can also result in this error."

170 ),

171 702: (

172 "This indicates that the device kernel took too long to execute. This can"

173 " only occur if timeouts are enabled - see the device attribute"

174 " ::CU_DEVICE_ATTRIBUTE_KERNEL_EXEC_TIMEOUT for more information."

175 " This leaves the process in an inconsistent state and any further CUDA work"

176 " will return the same error. To continue using CUDA, the process must be terminated"

177 " and relaunched."

178 ),

179 703: ("This error indicates a kernel launch that uses an incompatible texturing mode."),

180 704: (

181 "This error indicates that a call to ::cuCtxEnablePeerAccess() is"

182 " trying to re-enable peer access to a context which has already"

183 " had peer access to it enabled."

184 ),

185 705: (

186 "This error indicates that ::cuCtxDisablePeerAccess() is"

187 " trying to disable peer access which has not been enabled yet"

188 " via ::cuCtxEnablePeerAccess()."

189 ),

190 708: ("This error indicates that the primary context for the specified device has already been initialized."),

191 709: (

192 "This error indicates that the context current to the calling thread"

193 " has been destroyed using ::cuCtxDestroy, or is a primary context which"

194 " has not yet been initialized."

195 ),

196 710: (

197 "A device-side assert triggered during kernel execution. The context"

198 " cannot be used anymore, and must be destroyed. All existing device"

199 " memory allocations from this context are invalid and must be"

200 " reconstructed if the program is to continue using CUDA."

201 ),

202 711: (

203 "This error indicates that the hardware resources required to enable"

204 " peer access have been exhausted for one or more of the devices"

205 " passed to ::cuCtxEnablePeerAccess()."

206 ),

207 712: ("This error indicates that the memory range passed to ::cuMemHostRegister() has already been registered."),

208 713: (

209 "This error indicates that the pointer passed to ::cuMemHostUnregister()"

210 " does not correspond to any currently registered memory region."

211 ),

212 714: (

213 "While executing a kernel, the device encountered a stack error."

214 " This can be due to stack corruption or exceeding the stack size limit."

215 " This leaves the process in an inconsistent state and any further CUDA work"

216 " will return the same error. To continue using CUDA, the process must be terminated"

217 " and relaunched."

218 ),

219 715: (

220 "While executing a kernel, the device encountered an illegal instruction."

221 " This leaves the process in an inconsistent state and any further CUDA work"

222 " will return the same error. To continue using CUDA, the process must be terminated"

223 " and relaunched."

224 ),

225 716: (

226 "While executing a kernel, the device encountered a load or store instruction"

227 " on a memory address which is not aligned."

228 " This leaves the process in an inconsistent state and any further CUDA work"

229 " will return the same error. To continue using CUDA, the process must be terminated"

230 " and relaunched."

231 ),

232 717: (

233 "While executing a kernel, the device encountered an instruction"

234 " which can only operate on memory locations in certain address spaces"

235 " (global, shared, or local), but was supplied a memory address not"

236 " belonging to an allowed address space."

237 " This leaves the process in an inconsistent state and any further CUDA work"

238 " will return the same error. To continue using CUDA, the process must be terminated"

239 " and relaunched."

240 ),

241 718: (

242 "While executing a kernel, the device program counter wrapped its address space."

243 " This leaves the process in an inconsistent state and any further CUDA work"

244 " will return the same error. To continue using CUDA, the process must be terminated"

245 " and relaunched."

246 ),

247 719: (

248 "An exception occurred on the device while executing a kernel. Common"

249 " causes include dereferencing an invalid device pointer and accessing"

250 " out of bounds shared memory. Less common cases can be system specific - more"

251 " information about these cases can be found in the system specific user guide."

252 " This leaves the process in an inconsistent state and any further CUDA work"

253 " will return the same error. To continue using CUDA, the process must be terminated"

254 " and relaunched."

255 ),

256 720: (

257 "This error indicates that the number of blocks launched per grid for a kernel that was"

258 " launched via either ::cuLaunchCooperativeKernel or ::cuLaunchCooperativeKernelMultiDevice"

259 " exceeds the maximum number of blocks as allowed by ::cuOccupancyMaxActiveBlocksPerMultiprocessor"

260 " or ::cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags times the number of multiprocessors"

261 " as specified by the device attribute ::CU_DEVICE_ATTRIBUTE_MULTIPROCESSOR_COUNT."

262 ),

263 721: (

264 "An exception occurred on the device while exiting a kernel using tensor memory: the"

265 " tensor memory was not completely deallocated. This leaves the process in an inconsistent"

266 " state and any further CUDA work will return the same error. To continue using CUDA, the"

267 " process must be terminated and relaunched."

268 ),

269 800: "This error indicates that the attempted operation is not permitted.",

270 801: ("This error indicates that the attempted operation is not supported on the current system or device."),

271 802: (

272 "This error indicates that the system is not yet ready to start any CUDA"

273 " work. To continue using CUDA, verify the system configuration is in a"

274 " valid state and all required driver daemons are actively running."

275 " More information about this error can be found in the system specific"

276 " user guide."

277 ),

278 803: (

279 "This error indicates that there is a mismatch between the versions of"

280 " the display driver and the CUDA driver. Refer to the compatibility documentation"

281 " for supported versions."

282 ),

283 804: (

284 "This error indicates that the system was upgraded to run with forward compatibility"

285 " but the visible hardware detected by CUDA does not support this configuration."

286 " Refer to the compatibility documentation for the supported hardware matrix or ensure"

287 " that only supported hardware is visible during initialization via the CUDA_VISIBLE_DEVICES"

288 " environment variable."

289 ),

290 805: "This error indicates that the MPS client failed to connect to the MPS control daemon or the MPS server.",

291 806: "This error indicates that the remote procedural call between the MPS server and the MPS client failed.",

292 807: (

293 "This error indicates that the MPS server is not ready to accept new MPS client requests."

294 " This error can be returned when the MPS server is in the process of recovering from a fatal failure."

295 ),

296 808: "This error indicates that the hardware resources required to create MPS client have been exhausted.",

297 809: "This error indicates the the hardware resources required to support device connections have been exhausted.",

298 810: "This error indicates that the MPS client has been terminated by the server. To continue using CUDA, the process must be terminated and relaunched.",

299 811: "This error indicates that the module is using CUDA Dynamic Parallelism, but the current configuration, like MPS, does not support it.",

300 812: "This error indicates that a module contains an unsupported interaction between different versions of CUDA Dynamic Parallelism.",

301 900: ("This error indicates that the operation is not permitted when the stream is capturing."),

302 901: (

303 "This error indicates that the current capture sequence on the stream"

304 " has been invalidated due to a previous error."

305 ),

306 902: (

307 "This error indicates that the operation would have resulted in a merge of two independent capture sequences."

308 ),

309 903: "This error indicates that the capture was not initiated in this stream.",

310 904: ("This error indicates that the capture sequence contains a fork that was not joined to the primary stream."),

311 905: (

312 "This error indicates that a dependency would have been created which"

313 " crosses the capture sequence boundary. Only implicit in-stream ordering"

314 " dependencies are allowed to cross the boundary."

315 ),

316 906: ("This error indicates a disallowed implicit dependency on a current capture sequence from cudaStreamLegacy."),

317 907: (

318 "This error indicates that the operation is not permitted on an event which"

319 " was last recorded in a capturing stream."

320 ),

321 908: (

322 "A stream capture sequence not initiated with the ::CU_STREAM_CAPTURE_MODE_RELAXED"

323 " argument to ::cuStreamBeginCapture was passed to ::cuStreamEndCapture in a"

324 " different thread."

325 ),

326 909: "This error indicates that the timeout specified for the wait operation has lapsed.",

327 910: (

328 "This error indicates that the graph update was not performed because it included"

329 " changes which violated constraints specific to instantiated graph update."

330 ),

331 911: (

332 "This indicates that an async error has occurred in a device outside of CUDA."

333 " If CUDA was waiting for an external device's signal before consuming shared data,"

334 " the external device signaled an error indicating that the data is not valid for"

335 " consumption. This leaves the process in an inconsistent state and any further CUDA"

336 " work will return the same error. To continue using CUDA, the process must be"

337 " terminated and relaunched."

338 ),

339 912: "Indicates a kernel launch error due to cluster misconfiguration.",

340 913: ("Indiciates a function handle is not loaded when calling an API that requires a loaded function."),

341 914: ("This error indicates one or more resources passed in are not valid resource types for the operation."),

342 915: ("This error indicates one or more resources are insufficient or non-applicable for the operation."),

343 916: ("This error indicates that an error happened during the key rotation sequence."),

344 917: (

345 "This error indicates that the requested operation is not permitted because the"

346 " stream is in a detached state. This can occur if the green context associated"

347 " with the stream has been destroyed, limiting the stream's operational capabilities."

348 ),

349 999: "This indicates that an unknown internal error has occurred.",

350}

Coverage for cuda / core / _utils / driver_cu_result_explanations_frozen.py: 0.00%

1 statements