What's New
CUPTI contains a number of changes and new features as part
of the CUDA Toolkit 6.0 release.
- Two new CUPTI activity kinds have been introduced to enable two new types of source-correlated data collection. The Instruction Execution kind collects SASS-level instruction execution counts, divergence data, and predication data. The Shared Access kind collects source correlated data indication inefficient shared memory accesses.
- CUPTI now provides support for CUDA applications using Unified Memory. A new activity record reports Unified Memory activity such as transfers to and from a GPU and the number of Unified Memory related page faults.
- CUPTI now recognized and reports the special MPS context that is used by CUDA applications running on a system with MPS enabled.
- The CUpti_ActivityContext activity record has been updated in a backwards compatible manner. See Context Activity Record for more information.
- To ensure that metric values are calculated as accurately as possible, a new metric API is introduced. Function cuptiMetricGetRequiredEventGroupSets can be used to get the groups of events that should be collected at the same time.
- Execution overheads introduced by CUPTI have been dramatically decreased.
- The new activity buffer API introduced in CUDA Toolkit 5.5 is now required. The legacy cuptiActivityEnqueueBuffer and cuptiActivityDequeueBuffer functions have been removed.
Table of Contents
- 1. Usage
- 2. Modules
- 3. Data Structures
- 3.1. CUpti_Activity
- 3.2. CUpti_ActivityAPI
- 3.3. CUpti_ActivityBranch
- 3.4. CUpti_ActivityCdpKernel
- 3.5. CUpti_ActivityContext
- 3.6. CUpti_ActivityDevice
- 3.7. CUpti_ActivityDeviceAttribute
- 3.8. CUpti_ActivityEnvironment
- 3.9. CUpti_ActivityEvent
- 3.10. CUpti_ActivityEventInstance
- 3.11. CUpti_ActivityFunction
- 3.12. CUpti_ActivityGlobalAccess
- 3.13. CUpti_ActivityInstructionExecution
- 3.14. CUpti_ActivityKernel
- 3.15. CUpti_ActivityKernel2
- 3.16. CUpti_ActivityMarker
- 3.17. CUpti_ActivityMarkerData
- 3.18. CUpti_ActivityMemcpy
- 3.19. CUpti_ActivityMemcpy2
- 3.20. CUpti_ActivityMemset
- 3.21. CUpti_ActivityMetric
- 3.22. CUpti_ActivityMetricInstance
- 3.23. CUpti_ActivityModule
- 3.24. CUpti_ActivityName
- 3.25. CUpti_ActivityObjectKindId
- 3.26. CUpti_ActivityOverhead
- 3.27. CUpti_ActivityPreemption
- 3.28. CUpti_ActivitySharedAccess
- 3.29. CUpti_ActivitySourceLocator
- 3.30. CUpti_ActivityUnifiedMemoryCounter
- 3.31. CUpti_ActivityUnifiedMemoryCounterConfig
- 3.32. CUpti_CallbackData
- 3.33. CUpti_EventGroupSet
- 3.34. CUpti_EventGroupSets
- 3.35. CUpti_MetricValue
- 3.36. CUpti_ModuleResourceData
- 3.37. CUpti_NvtxData
- 3.38. CUpti_ResourceData
- 3.39. CUpti_SynchronizeData
- 4. Data Fields
- Notices