NVIDIA CUDA Profiling Tools Interface (CUPTI) - CUDA Toolkit
The NVIDIA® CUDA Profiling Tools Interface (CUPTI) is a dynamic library that enables the creation of profiling and tracing tools that target CUDA applications. CUPTI provides a set of APIs targeted at ISVs creating profilers and other performance optimization tools:
- the Activity API,
- the Callback API,
- the Event API,
- the Metric API,
- the Profiler API,
- the PC Sampling API, and
- the Checkpoint API
Using these CUPTI APIs, independent software developers can create profiling tools that provide low and deterministic profiling overhead on the target system, while giving insight into the CPU and GPU behavior of CUDA applications. Normally packaged with the CUDA Toolkit, NVIDIA occasionally uses this page to provide CUPTI improvements and bug fixes between toolkit releases.
- Trace CUDA API by registering callbacks for API calls of interest
- Full support for entry and exit points in the CUDA C Runtime (CUDART) and CUDA Driver
- GPU workload trace for the activities happening on the GPU, which includes kernel executions, memory operations (e.g., Host-to-Device memory copies) and memset operations.
- CUDA Unified Memory trace for transfers from host to device, device to host, device to device and page faults on CPU and GPU etc.
- Normalized timestamps for CPU and GPU trace
- Profile hardware and software event counters, including:
- Utilization metrics for various hardware units
- Instruction count and throughput
- Memory load/store events and throughput
- Cache hits/misses
- Branches and divergent branches
- Many more
- Enables automated bottleneck identification based on metrics such as instruction throughput, memory throughput, and more
- Range profiling to enable metric collection over concurrent kernel launches within a range
- Metrics attribution at the high-level source code and the executed assembly instructions.
- Device-wide sampling of the program counter (PC). The PC Sampling gives the number of samples for each source and assembly line with various stall reasons.
- CUPTI has made the following changes as part of the CUDA Toolkit 11.6 release:
- Two new fields channelID and channelType are added in the activity records for kernel, memcpy, peer-to-peer memcpy and memset to output the ID and type of the hardware channel on which these activities happen. Activity records
CUpti_ActivityMemset3are deprecated and replaced by new activity records
- New fields isMigEnabled, gpuInstanceId, computeInstanceId and migUuid are added in the device activity record to provide MIG information for the MIG enabled GPU. Activity record CUpti__ActivityDevice3 is deprecated and replaced by a new activity record CUpti__ActivityDevice4.
- A new field utilizedSize is added in the memory pool activity record to provide the utilized size of the memory pool. Activity record CUpti_ActivityMemoryPool is deprecated and replaced by a new activity record CUpti_ActivityMemoryPool2.
- API cuptiActivityRegisterTimestampCallback and callback function CUpti_TimestampCallbackFunc are added to register a callback function to obtain timestamp of user's choice instead of using CUPTI provided timestamp in activity records.
- Profiling API supports profiling of the OptiX application.
- Linux x86_64
- Windows x86_64
- Linux PowerPC
- Linux aarch64 SBSA
- DRIVE OS QNX aarch64
- DRIVE OS Linux aarch64
 available in the Embedded or Drive toolkits only
Supported NVIDIA GPU architectures
- Activity and Callback APIs
- All architectures supported by CUDA Toolkit
- Event and Metric APIs
- Profiling and PC Sampling APIs
- Ampere: A100 with Multi-Instance GPU, GA10x
- Volta: GV100, GV10B
- CUPTI can be found in the CUDA Toolkit 11.6 production release
- Please use the following drivers
- 511.23 (Windows)
- 510.39.01 (Linux)
To provide feedback, request additional features, or report issues, please use the Developer Forums.
When installing CUDA Toolkit 11.6 and specifying options, be sure to select CUDA > Development > Tools > CUPTI.