• <xmp id="om0om">
  • <table id="om0om"><noscript id="om0om"></noscript></table>
  • 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.

    There is currently no CUPTI update to the CUDA Toolkit 12.2. You may obtain the latest version of CUPTI by Downloading the CUDA Toolkit 12.2

      Download the CUDA Toolkit 12.2 Now 
    Revision History

    Key Features

    • 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.
    See the CUPTI User Guide for a complete listing of hardware and software event counters available for performance analysis tools.

    Updates in CUDA Toolkit 12.2

      New Features

      • A new set of CUPTI APIs for collection of SASS metric data at the source level are provided in the header file cupti_sass_metrics.h. These support a larger set of metrics compared to the CUPTI Activity APIs for source-level analysis. SASS to source correlation can be done in the offline mode, similar to the PC sampling APIs. Hence the runtime overhead during data collection is lower. Refer to the section CUPTI SASS Metrics API for more details. Please note that this is a Beta feature, interface and functionality are subject to change in a future release.
      • CUPTI now reports fatal errors, non-fatal errors and warnings instantaneously through callbacks. A new callback domain CUPTI_CB_DOMAIN_STATE is added for subscribing to the instantaneous error reporting. Corresponding callback ids are provided in the struct CUpti_CallbackIdState.
      • Added support for profiling of device graphs and host graphs that launch device graphs. There are some known limitations, please refer to the Known Issues section for details.
      • Change in the stream attribute value is communicated by issuing the resource callback. Refer to the struct CUpti_StreamAttrData and callback id CUPTI_CBID_RESOURCE_STREAM_ATTRIBUTE_CHANGED added in the enum CUpti_CallbackIdResource.
      • New API cuptiGetErrorMessage provides descriptive message for CUPTI error codes.
      • Removed the deprecated API cuptiDeviceGetTimestamp from the header cupti_events.h.


    Supported platforms

    • Linux x86_64[1]
    • Windows x86_64[1]
    • Linux PowerPC[1]
    • Linux aarch64 SBSA[1]
    • DRIVE OS QNX aarch64[2]
    • DRIVE OS Linux aarch64[2]
         [1] available in the CUDA Desktop Toolkit only
         [2] 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
      • Volta
      • Pascal
      • Maxwell
    • Profiling and PC Sampling APIs
      • Hopper: GH100
      • Ada: AD10x
      • Ampere: A100 with Multi-Instance GPU, GA10x
      • Turing
      • Volta: GV100, GV10B

    CUDA Toolkit


      Please use the following drivers
      • 536.23 (Windows)
      • 535.54.03 (Linux)
      provided with CUDA Toolkit 12.2 production release or a more recent version.



    To provide feedback, request additional features, or report issues, please use the Developer Forums.

    Installation Overview

    When installing CUDA Toolkit 12.2 and specifying options, be sure to select CUDA > Development > Tools > CUPTI.