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, and
- the Profiler 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. CUPTI is delivered as a dynamic library on all platforms supported by CUDA.
- Trace CUDA API usage by registering callbacks for API calls of interest
- Full support for entry and exit points in the CUDA C Runtime (CUDART) and CUDA Driver
- Trace for applications using CUDA Unified Memory, OpenACC directives, OpenMP directives and NVTX annotations
- Sample hardware and software event counters, including:
- Instruction count and throughput
- Memory load/store events and throughput
- Cache hits/misses
- Branches and divergent branches
- NVLink transfers
- Many more
- Enables bottleneck identification based on metrics such as instruction throughput, memory throughput, and more
- o Ability to collect metrics at the assembly instruction (SASS) level and support to correlate lines between executed assembly instructions and high-level source code
- PC sampling feature collects samples of the program counter (PC) and shows the stall reasons at source level
- Normalized timestamps for CPU and GPU events
As part of the CUDA Toolkit 11.1 release,
- CUPTI adds profiling support for the new NVIDIA Ampere GPUs with compute capability 8.6.
- Added a new field graphId in the activity records for kernel, memcpy, peer-to-peer memcpy and memset to output the unique ID of the CUDA graph that launches the activity through CUDA graph APIs. To accommodate this change, activity records CUpti_ActivityMemcpy3, CUpti_ActivityMemcpyPtoP2 and CUpti_ActivityMemset2 are deprecated and replaced by new activity records CUpti_ActivityMemcpy4, CUpti_ActivityMemcpyPtoP3 and CUpti_ActivityMemset3. And kernel activity record CUpti_ActivityKernel5 replaces the padding field with graphId. Added a new API cuptiGetGraphId to query the unique ID of the CUDA graph.
- Added a new API cuptiActivityFlushPeriod to set the flush period for the worker thread.
- Added support for profiling cooperative kernels using Profiling APIs.
- Added NVLink performance metrics (nvlrx__* and nvltx__*) using the Profiling APIs. These metrics are available on devices with compute capability 7.0, 7.5 and 8.0, and these can be collected at the context level. Refer to the table Metrics Mapping Table for mapping between earlier CUPTI metrics and the Perfworks NVLink metrics for devices with compute capability 7.0.
- Linux x86_64
- Windows x86_64
- Linux PowerPC
- Linux aarch64 SBSA
- DRIVE OS QNX aarch64
- DRIVE OS Linux aarch64
- Android aarch64
- Activity and Callback APIs
- All architectures supported by CUDA Toolkit
- Event and Metric APIs
- Profiling APIs
- CUPTI can be found in the CUDA Toolkit 11.1 production release
- 456.33 (Windows)
- 455.23.02 (Linux)
 available in the Embedded or Drive toolkits only
 does not support Profiling API data collection
 MacOS is supported as a host, but not a target.
Supported NVIDIA GPU architectures
Please use the following drivers
To provide feedback, request additional features, or report issues, please use the Developer Forums.
When installing CUDA Toolkit 11.1 and specifying options, be sure to select CUDA > Development > Tools > CUPTI.