# 1\. Release Notes [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#release-notes "Permalink to this headline") Nsight Compute Release Notes. Release notes, including new features and important bug fixes. Supported platforms and GPUs. List of known issues for the current release. ## 1.1. Release Notes [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#id1 "Permalink to this headline") ### 1.1.1. Updates in 2026.1.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2026-1-1 "Permalink to this headline") **Resolved Issues** - Fixed an issue where metric colors in a timeline row could differ from those shown in the corresponding tooltip. - Fixed an incorrect calculation in the Roofline Analysis rule that prevented it from triggering in all expected cases. - Fixed an issue where the High Pipe Utilization rule might not analyze all pipelines. ### 1.1.2. Updates in 2026.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2026-1 "Permalink to this headline") **General** - Added support for CUDA 13.2. - The minimum supported version of macOS is now 13.0. - The [Python Report Interface](https://docs.nvidia.com/nsight-compute/PythonReportInterface/index.html.md#introduction) can now be installed as a standalone package from PyPI using `pip install ncu-report`. **NVIDIA Nsight Compute** - Nsight Compute now supports profiling Linux (aarch64 sbsa) targets from macOS hosts. - The [Start Activity Dialog](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#start-activity-dialog) now support editing multi-line command line arguments. - Added the [Report Merge Tool](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#report-merge-tool). It enables you to combine multiple reports into one. It is particularly useful for multi-GPU systems and scenarios when comparing and analyzing several reports individually becomes impractical. - Added the [Clustering Window](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#clustering-window). It helps you analyze and compare multiple profiling reports by grouping similar reports together. This makes it easier to identify performance patterns and find relationships between different profiling sessions. - Added [Register Dependencies](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-additional-tables) analysis to the Source page. It helps you to identify general purpose register dependencies and occupancy issues due to live register pressure. Added _Attributed Live Registers_ and _Output Registers_ metric columns in the Source page views. Renamed _Instructions & Scoreboards_ to _Instructions & Dependencies_. - The [Source](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#pre-defined-source-metrics) page now shows the [metric pipelines](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#pipelines) associated with each instruction. - Added a [CUDA Graph Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#cuda-graph-viewer) tool window that dynamically visualizes CUDA Graphs during interactive profiling. - [Timelines](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#timelines) on the Details page now overlay related metrics in a single row when possible. They now also show a max bar in the background to indicate the maximum value at any zoom level. Rows can be switched between theoretical peak and collected maximum value for the Y-axis scale. - The Instruction Statistics section now has thread-level charts. - Improved alignment of rule elements on the Details page. - Moved the section body dropdowns to the left of the Details page section headers. - Renamed “Kernel Analysis” to “SASS Analysis” in Options dialog. - The Save As dialog now supports the `.ncu-repz` file extension for compressed reports. **NVIDIA Nsight Compute CLI** - [Mandatory concurrent kernels](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#mandatory-concurrent-kernels) (e.g. NCCL communication kernels) can now be profiled across processes from the same process tree using the `--communicator shmem` option. - The tool now generates a persisted log file in case of non-recoverable errors. - The default value of `--clock-control` is now `boost`. **Resolved Issues** - Fixed issues with OptiX command lists in interactive profiling mode. - Added `occupancy` as an option to `--query-metrics-collection`. - Fixed issues with the Active Clusters graph in the [occupancy calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#graphs). - Fixed that the inline functions table could show incorrect metrics values in some cases. - Fixed issues with syntax highlighting on the Souce page. - Improved the performance of the [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) view for large SASS diffs. - Fixed issues with alignment of multi-pass PM sampling [timelines](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#timelines) without context switch trace. - Using an unknown metric in a PM sampling timeline section file is now an error. - Fixed issues with opcode category tooltips in the Instruction Statistics charts. - Fixed issues in node-level profiling of CUDA device launchable graphs. ### 1.1.3. Updates in 2025.4.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-4-1 "Permalink to this headline") **General** - Added support for profiling OptiX workloads with the interactive profile activity. - The Resources tool window now shows a _CUDA ID_ column for graph resources. **Resolved Issues** - Fixed issues in the Report Merge Tool. - Fixed issues in the Clustering Window. - Fixed an issue with the formatting of entries in the Inline Functions table. - Fixed an issue that could cause the UI to crash when switching between kernels in the Source page in certain cases. - Fixed an issue that the `ncu` CUDA Toolkit script may fail if unrecognized versions of Nsight Compute are installed separately. - Fixed several incorrect metrics in the `PmSampling.section`. - Fixed an issue that PM Sampling timelines could be distorted due to broken initial samples. - Fixed a crash with node-level graph profiling in app-range replay mode. If graphs are uploaded outside the range, an error indicating that profiling it is not supported is now shown. - Fixed a crash when instantiating device-side graphs with memory nodes. ### 1.1.4. Updates in 2025.4 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-4 "Permalink to this headline") **General** - Added support for profiling CUDA tile workloads. - Introduced a new Tile section to summarize tile dimensions and pipeline utilization, displayed when enabled and a tile workload is profiled. - Source page supports correlation between SASS and high-level Tile code (limited to cuTile Python code). - Added a new `ncu-repz` file format for [zstd](https://facebook.github.io/zstd/) compressed report files. - Added support for locking GPUs to boost clock instead of base on Ampere and newer GPU. Use the `boost` and `force-boost` [options](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-profile) on supported drivers. - [Warp sampling](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#warp-sampling) by default now focuses on the _Not Issued_ (`(_not_issued)`) variants of the metrics. This is to avoid pointing to source locations where warp stalls are mitigated by having sufficient numbers of warps during an issue cycle to hide latency. - Added support for node-level profiling of CUDA conditional graphs, including device-updatable nodes and nodes that can set conditional graph handles. - Added support for node-level profiling of CUDA graphs launched from the device (DGL), including host graph nodes that can launch DGL. - Source page now displays [symbol labels](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-navigation): A new column for symbol labels has been added, and symbol labels are shown alongside addresses in SASS instruction disassembly. This change aligns the output with that of the nvdisasm tool. - Added support for collecting Warp sampling metrics with PM sampling allowing user to see function-level warp stalls for the selected time range in the timeline. See the [Function Stats tool window](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-function-stats) for details. ### 1.1.5. Updates in 2025.3.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-3-1 "Permalink to this headline") **General** - Improved the charts in the Compute Workload Analysis section to better distinguish between per\_cycle\_active and per\_cycle\_elapsed metrics. **Resolved Issues** - Fixed an issue where kernels using the compile-time attribute `__block_size__` were launched with incorrect grid dimensions. - Fixed an issue with timline y-axis labels being showing unexpected units for small max values. - Fixed a crash when stepping applications in interactive profiling mode. - Fix that roofline charts missed showing achived value data in some cases. - Fixed that duplicated tooltips could be shown for some links in the Memory Chart. - Fixed a potential hang when setting `--pm-sampling-buffer-size` to very large values. - Fixed several rules to not show non-actionable warnings for unsupported, missing metrics when profiling on mobile chips. ### 1.1.6. Updates in 2025.3 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-3 "Permalink to this headline") **General** - Added support for CUDA 13.0. See the tool’s CUDA driver [system requirements](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html.md#system-requirements). - Added or improved support for Blackwell chips. - For Green Context launches, `launch__waves_per_multiprocessor` is now scaled to the number of SMs in the Green Context. - Added support for profiling individual nodes of device-launchable CUDA graphs launched from the host. - Added metric `launch__persisting_l2_cache_size` to the Memory Workload Analysis section. - Removed metric `profiler__pmsampler_dropped_samples`. - Added [support](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-profile) for not importing SASS cubins into the report. **NVIDIA Nsight Compute** - The [Source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#pre-defined-source-metrics) now shows the instruction category in SASS and the instruction mix for high-level source. - Added a new [instruction mix and scoreboard dependencies table](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-additional-tables) to the Source page. - Added improved tooltips to the [memory chart](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#memory-chart). - Added information on the GPC Constant Cache (GCC) and DSMEM atomics in the [memory tables](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#memory-tables). - The [Metric Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-metric-details) tool window now shows the breakdown for [throughput metrics](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#metrics-entities). - Added support for searching web forum and rule results. - Multiple results from the same search source are now combined to make the output more readable. - Improved the [occupancy calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#occupancy-calculator). **NVIDIA Nsight Compute CLI** - Added the option [–forward-signals](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-launch) to transparently forward signals to the profiled application. **Resolved Issues** - Fixed that some `ncu` console messages were truncated after 1024 characters. - Fixed some display issues related to Green Context tables. - Improved the performance of remote profiling in application replay mode. - Fixed a hang in certain scenarios when profiling dependent kernels with device-mapped host allocations. - Fixed missing correlation between JIT-compiled PTX to SASS in some situations. - Fixed an error when profiling a CUDA graph kernel node doing a cluster launch on driver 580 or newer. ### 1.1.7. Updates in 2025.2.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-2-1 "Permalink to this headline") **Resolved Issues** - Updated OpenSSL to 3.0.16 - Fixed an issue that caused the Device Memory table of the Memory Workload Analysis section to show up empty for chips of type GB100 and GB102 ### 1.1.8. Updates in 2021.2.9 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-9 "Permalink to this headline") **NVIDIA Nsight Compute** - Clarify when not all metrics for the roofline chart could be collected on the current chip. ### 1.1.9. Older Versions [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#older-versions "Permalink to this headline") #### Updates in 2025.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-2 "Permalink to this headline") **General** > - Added support for collecting C2C link information on Blackwell GPUs. > > - [CPU call stack filtering](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#cpu-stack-filtering) now supports Python call stacks. > > - Instruction statistics now show warp- and thread-level instruction counts per opcode category. > Added new metrics `sass__inst_executed_per_opcode_category` and `sass__thread_inst_executed_per_opcode_category`. > See the [Metrics Reference](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#metrics-reference) for details. > > - Enhanced several rules to produce tables pointing to the source location of interest. > > - Improved the NvRules API to support [generic tables](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md#NvRules.IFrontend.generate_table) for the UI and CLI. > > - Improved the [NvRules](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md) and [Python Report Interface](https://docs.nvidia.com/nsight-compute/PythonReportInterface/index.html.md) documentations to be more pythonic. > > - Added APIs to the Python Report Interface for querying rules and source markers in the report. > > - Added [Occupancy Calculator Python Interface](https://docs.nvidia.com/nsight-compute/OccupancyCalculatorPythonInterface/index.html.md#introduction), which provides a Python-based interface for performing occupancy calculations and analysis of kernels on NVIDIA GPUs. **NVIDIA Nsight Compute** - Added product-wide [search](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-search) functionality via a new search bar and tool window. - The Source page now shows scoreboard dependencies in SASS. - Converted more tooltips into interactive tooltips. Interactive tooltips can now be pinned and dragged. - Added source correlation [navigation controls](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#navigation) which allow navigation to the previous or next block of correlated lines. **NVIDIA Nsight Compute CLI** - Added support for profiling [MPS applications](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#special-configurations-mps). - Added support for filtering kernels based on a [renamed kernels](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#kernel-renaming) configuration during profiling. **Resolved Issues** - CUDA Graphs in the [Resources View](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) use the current UI theme. - Resolved several issues when interacting with timelines on the Details page. - Resolved issues with Python syntax highlighting on the Source page. - Disabled deprecated columns in the API Stream tool window. - Fixed that the Source page may show incorrect correlation when some source files were not resolved. - Reduced the number of replay passes required for collecting the `PmSampling.section` on GH100 with applicable drivers. - Resolved that `--native-include` did not work properly when using range replay and `cu(da)ProfilerStop`. - Fixed an `Invalid or unsupported charset:ANSI_X3.4-1968` error when using the CLI on some systems. - Fixed that memory available for saving context state during replay may be computed incorrectly when the app was using managed memory. - Fixed that some metrics were not listed for collection in section files for GB20x GPUs. #### Updates in 2025.1.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-1-1 "Permalink to this headline") **General** - Added support for Optix 9.0 functions `optixClusterAccelComputeMemoryUsage` and `optixClusterAccelBuild`. **Resolved Issues** - Fixed a possible deadlock condition while handling the launch of child processes on Linux systems. - Fixed a possible crash of the Nsight Compute UI when switching to the Source Page. - Fixed the missing roofline ceilings in the Floating Point Operations Roofline for GB20x chips. #### Updates in 2025.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2025-1 "Permalink to this headline") **General** - All roofline sections are now included in the `full` section set. - [Range Replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay) and [app-range replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#application-range-replay) are now supporting the collection of instruction-level source metrics. - Rules are now supported for range replays. - Improved which launch metrics are available for ranges. - Added a new `launch__stack_size` metric in the _Launch Statistics_ section to report the configured stack size. - Added a new `sass__inst_executed_register_spilling` metric which counts the number of load and store instructions that were created by the compiler due to register spilling. - Nsight Compute now natively supports macOS arm64. **NVIDIA Nsight Compute** - Added interactive tooltips to [Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-details-page) and [Source](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) pages. An interactive tooltip can be used to compare different baselines. Its content can be copied to the clipboard using the copy icon button. - CUDA [Green Contexts support](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#green-contexts-support) is improved by showing TPC mask information in the Launch Statistics section, the [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window, and on the [Session page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#session-page). - Added heatmap to the [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) document to visualize the source code differences. - Added _Diff By_ drop down menu to the [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) document in the _SASS_ view, this allows you to choose the diff basis based on either _Opcode_ or _Full Instruction_. - Performance improvements in _SASS_ view. - The [Resources View](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) for CUDA Graphs can now visualize the graph structure directly in a new _Chart_ mode. - The [Memory Chart](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#memory-chart) now supports zoom and pan. - The [Metric Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-metric-details) tool window now shows PM sampling metrics from the timeline as context switched. - Improved the performance for deploying to target systems over [remote connections](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#remote-connections). **NVIDIA Nsight Compute CLI** - Added experimental support for profiling [MPS applications](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#special-configurations-mps). **Resolved Issues** - Fixed that on some systems, not all free GPU memory was considered when saving context memory for multi-pass data collection. - Fixed an incorrect multiplier in the calculation of non-tensor FP16 rooflines. - Fixed the metric `Avg. Threads Executed` for inlined functions with control flow. - Fixed that in some situations, no average was shown in the Source Statistics table for Warp Stall sampling metrics. - Fixed several SASS syntax highlighting issues. - Fixed an issue where the SM count wasn’t shown correctly in the report header when loading older reports. - Improved interactions between the [Metric Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-metric-details) tool window and the memory chart. #### Updates in 2024.4 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-4 "Permalink to this headline") **General** - Added support for the Blackwell architecture. - Added support for several `launch__*` metrics for CUDA graphs. - Added support for cuMemBatchDecompressAsync API in the [Range Replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay). **NVIDIA Nsight Compute** - A new feature overview is now shown the first time a new UI version is opened. - Switched the default orientation of the _Raw_ page to show metrics in rows and profile results in columns. - Added support for reporting register spilling compiler annotations on the Source page. - The [source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-navigation) has improved search with support for regular expression- and value-based lookups. - Added support to set a [Source View Profile](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-profiles) as the default profile to apply it automatically while opening a report. - Added hyperlinks for the line numbers and inline function addresses in the [Inline Table](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#additional-tables). This enabled you to quickly jump to the respective line number in the Source view and address in the SASS view. Added a new column Source File in the [Inline Table](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#additional-tables) to show the file name to which source belongs. - The [memory chart](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#memory-chart) can indicate or hide inactive elements. - Chart tooltips on the Details page now show more relevant information when a specific value is hovered. - [Roofline charts](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#roofline) now support showing the formula for ridge point calculation in the [metric details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-metric-details) tool window. - The [occupancy calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#occupancy-calculator) now considers the impact of block barriers for Hopper-architecture and newer GPUs. It also has improved controls to adjust input values. - The [remote connections](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#remote-connections) dialog now supports placeholders to deploy files to e.g. user-specific directories on the target system. **NVIDIA Nsight Compute CLI** - Added new `--nvtx-push-pop-scope` [command line option](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-profile) which allows to set push pop range scope process wide. **Resolved Issues** - Fixed UI scrolling issues on macOS trackpads. - Fixed that certain Python script errors were not properly reported when loading rule files. - On CUDA 12.7 drivers, [context switch trace](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#pm-sampling) can now filter events more precisely to the profiled CUDA context, even when profiling in containers. - [NVTX filtering](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#nvtx-filtering) now properly supports start/end ranges that start and end in different threads. - Fixed several issues with [range replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay-supported-apis) when capturing CUDA memcpy APIs. #### Updates in 2024.3.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-3-2 "Permalink to this headline") **NVIDIA Nsight Compute** - Added support for kernel profiling on the Microsoft Compute Driver Model (MCDM). Requires an NVIDIA display driver version of 565 or higher. **Resolved Issues** - Fixed the calculation of peak throughput values for select pipelines on chips using the Ada GPU architecture, including the NVIDIA L20 and NVIDIA L40 data-center GPUs. - Re-enabled the target-side data collection for the [Acceleration Structure Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#acceleration-structure-viewer) on Linux (aarch64 and aarch64 sbsa), which was disabled in 2024.3.1. #### Updates in 2024.3.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-3-1 "Permalink to this headline") **NVIDIA Nsight Compute** - Added the ability to use the ESC button to close the search popup in the views of the Source Page. - Disabled the [Acceleration Structure Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#acceleration-structure-viewer) on Linux (aarch64 and aarch64 sbsa) as the rendering view may cause a crash of the Nsight Compute UI. The functionality will be restored in a future version. **NVIDIA Nsight Compute CLI** - Added back the help documentation for command line parameters `--set` and `--list-sets`. **Resolved Issues** - Fixed an issue with PM Sampling on Hopper GPUs that could cause the target application to hang during profiling. - Fixed an issue that caused CPU stack frames to not show up if stack frames could not be fully resolved. - Fixed an issue with the heights of the tables in the Metric Details window. - Fixed a possible crash of the Nsight Compute UI while preparing the information for the Source Page. - Fixed an issue that caused a failure when the request to save a partial profile report is canceled. #### Updates in 2024.3 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-3 "Permalink to this headline") **NVIDIA Nsight Compute** - Added syntax highlighting support for section and rule files when open these as a new document. These documents now support edit, save (ctrl + s), text search (ctrl + f), zoom in (ctrl + mouse scroll down), zoom out (ctrl + mouse scroll up) functionalities. - In [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) document, added synchronized horizontal scrolling support across both sides for frozen columns. - Added hyperlinks for section and rule files in the [Metric Selection](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#metric-selection) tool window. - Improved Inline Table source syntax highlighting. - The ‘Connect’ button has been renamed to ‘Start Activity’. - The ‘Connection Dialog’ has been renamed ‘Start Activity’. - Improved Progress Log window in [Start Activity Dialog](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-dialog) to show progress status of files being deployed to the remote host machine. - Added file menu options to Save Filtered Results and Save Selected Results from an opened report to a new report. Also added context menu option Save Results to save the selected results from the Summary Table or Raw Table to a new report. **NVIDIA Nsight Compute CLI** - Added support for exporting filtered results from an imported profile report to a new file using the –export and –import options together. - The –kernel-id option now supports regex for specifying the context id and stream id. #### Updates in 2024.2.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-2-1 "Permalink to this headline") **General** - Improved performance when filtering by NVTX context in kernel and application replay. - Improved documentation for metric units and terms. **NVIDIA Nsight Compute** - Improved tooltips for the memory chart. - Improved timeline row maximum selection for PM sampling metrics. **Resolved Issues** - Fixed an issue that the report result dropdown may not update when some options were changed. - Fixed an issue with the Source page statistics table. - Fixed an issue with PM sampling reporting incomplete data. - Fixed an issue with filering using NVTX start/end ranges. - Fixed an issue with demangling Numba CUDA kernel names. - Fixed an issue with profiling multi-ctx applications on vGPU. - Fixed an issue that resulted in L1 caches not always being invalidated for every pass. - Fixed an issue with applications using `execlp`. #### Updates in 2024.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-2 "Permalink to this headline") **General** - PM sampling timelines now show the sampled GPU workload activities. - Added support for collecting [Python Call Stacks](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-nvtx-page) alongside native ones to better understand the context of a workload in Python applications. - Demangled kernel names can now be automatically simplified or manually renamed. For reports with multiple results, all names are considered during simplification to make them easier to distinguish. - Removed support for ppc64le. **NVIDIA Nsight Compute** - Redesigned the [report header](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-header) for easier access to all report pages. All actions are now sorted into clearly labeled buttons. The focused result selection was integrated into the current row. When adding the current result as a baseline, the row itself is updated to reflect this, instead of showing a separate one. - Redesigned [Source Page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) and [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) controls to allow more vertical space. - The [Source Page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) and [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) Navigate By dropdowns can be linked to the respective other view. Changing column names from one dropdown will change it in the other view, too. This applies only if a column is also available in the second view. - Added Inline Table support in the [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) document to each side separately. - Added rich support for Python and Fortran source syntax highlighting. Enhanced CUDA-C and PTX syntax highlighting. - Added a Statistics Table to the Source Page that allows you to quickly see aggregated metrics across a custom selection of lines. - Improved tooltips in the memory chart to show more detailed information when metrics are missing. - The [Acceleration Structure Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#acceleration-structure-viewer) can now compute ray-geometry intersection and traversal timing heatmaps. - Added support for ignoring directories in the section search folder. - Added support for specifying [custom metric descriptions](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#custom-descriptions) in section files. - Added a warning if the opened report is newer than the UI and may not be fully compatible. **NVIDIA Nsight Compute CLI** - The raw page csv output now includes metric instance values when these enabled for printing. **Resolved Issues** - Improved handling of short workloads during PM sampling. - Improved units for several metrics. - Fixed an issue that some metrics did not show aggregates on the Summary and Raw pages. - Fixed an issue that profiled applications could inadvertently overwrite which PerfWorks library is loaded by the tool. - Fixed an issue that kernel names including $ were modified by the shell when profiling them from the System Trace activity. - Fixed an issue that reports could not be saved to the expected file extension in certain cases. #### Updates in 2024.1.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-1-1 "Permalink to this headline") **General** - Added clarification that for profiling a range with multiple, active CUDA Green Contexts, counter values that are not attributable to SMs will be aggregated over all these Green Contexts. **Resolved Issues** - Changed the way the PerfWorks library is loaded into the target application’s process space. This addresses possible connection errors in case the library search path includes other directories with PerfWorks libraries. - Fixed an issue that caused PM sampling data to be missing from the results of a Profile Series. - Fixed the incorrect calculation of the percentage values in the Inline Function table. - Fixed a potential crash of the NVIDIA Nsight Compute UI when PM sampling data was requested, but no sample was collected. #### Updates in 2024.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2024-1 "Permalink to this headline") **General** - Switched to using OpenSSL version 3.0.10. - Added new metrics available when profiling on CUDA Green Contexts. - Reduced the number of passes required for collecting [PM sampling](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#pm-sampling) sections. - Counter domains can now be specified for PM sampling metrics in section files. - PM sampling metrics can now be queried in the command line and Metric Details window by specifying the respective `collection` option. - Added a new optional [PmSampling\_WarpStates](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#sections-and-rules) section for understanding warp stall reasons over the workload duration. - Added a new rule for detecting load imbalances. - Improved the performance of graph-level profiling on new drivers. - Updated the [metrics compatibility](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#compatibility) table for OptiX cmdlists and instruction-level SASS metrics. - Improved PM sampling results by dynamically recollecting metrics to avoid outlier pass groups. Use the new [pm-sampling-max-passes](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#pm-sampling) option to control the maximum number of dynamically replayed passes. - Added _interKernelCommunication_ sample CUDA application to show how to use NVIDIA Nsight Compute to profile kernels that depend on each other and must be launched concurrently. Refer to the `README.TXT` file and sample code under `extras/samples/interKernelCommunication`. **NVIDIA Nsight Compute** - Added SASS view and Source Markers support in [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison). - Improved [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison) diff visualization by adding empty lines on other side of inserted/deleted lines. - The Source page [column chooser](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-metrics) can now be opened directly from the Navigation drop down. - Added support to update a [source-page profile](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiles). - Added a [Launch Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-launch-details) tool window for showing information about individual launches within larger workloads like OptiX command lists. - Added support for CUDA Green Contexts in the [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window, the Launch Statistics section and the report header. - Property metrics can now be queried in the Metric Details window. - Added support to show if a CUDA Graph kernel node is device-side updatable in the [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window. **NVIDIA Nsight Compute CLI** - Improved documentation on NVTX expressions and command line output when a potentially incorrect expression led to no workloads being profiled. - Improved checking for invalid expressions when using the `--target-processes-filer` option. **Resolved Issues** - Fixed that the L1 cache achieved roofline value was missing when profiling on GH100. - Fixed several “Launch Failed” errors when collecting instruction-level SASS metrics. - Fixed that [Live Register](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-metrics) values would be too high for some workloads. - Fixed a scrolling issue on the Source page when collapsing a multi-file view. - Fixed an issue that no PM sampling data would be shown in the timeline when context switch trace was not available. - Fixed a display issue in the memory chart when adding baselines. - Fixed a crash when adding baselines. - Fixed a crash in timeline views when not all configured data was available. - Fixed that the application history was not always deleted when selecting Reset Application Data. - Fixed an error in the metric compatibility documentation. #### Updates in 2023.3.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-3-1 "Permalink to this headline") **General** - Switched to using OpenSSL version 1.1.1w. - Improved the speedup estimates for rule IssueSlotUtilization as well as its child rules. - Updated report files and documentation for the samples located at `extras/samples/`. **Resolved Issues** - Fixed collection of context switch data during [PM Sampling](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#pm-sampling) when using [Range Replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay). - Fixed potential crash of NVIDIA Nsight Compute when an invalid regular expression was provided as requested metric. - Improved the performance of NVIDIA Nsight Compute in cases where only a single process is being profiled and `--target-processes all` was specified. - Fixed an issue of reporting too high register counts on the Source Page. - Fixed a bug that could cause a GPU fault while collecting SW counters through PerfWorks. - Fixed showing incorrect baseline values for the Runtime Improvement values on the Summary Page. #### Updates in 2023.3 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-3 "Permalink to this headline") **General** - NVIDIA Nsight Compute now supports collecting many metrics by [sampling the GPU’s performance monitors (PM)](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#pm-sampling) periodically at fixed intervals. The results can be visualized on a [timeline](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#details-page). - Added WSL profiling support on Windows 10 WSL with OS build version 19044 and greater. WSL profiling is not supported on Windows 10 WSL for systems that exceed 1 TB of system memory. - Rule outputs are prioritized to improve the accuracy of estimated speedups. The [Summary](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#summary-page) page now shows the most actionable optimization advices when a result row is selected. - Improved the handling and reporting for unavailable metrics during collection and when applying rules. - Added _instructionMix_ sample CUDA application and document to show how to use NVIDIA Nsight Compute to analyze and identify the performance bottleneck due to an imbalanced instruction mix. Refer to the `README.TXT` file, sample code, and document under `extras/samples/instructionMix`. **NVIDIA Nsight Compute** - Added support to see the source files of two profile results side by side using [Source Comparison](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#source-comparison). This allows you to quickly identify source differences and understand changes in metric values. - The [Summary](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#summary-page) page is now the default page when a report is opened. Previous behavior can be enabled in the [options](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profile) dialog. - On the [Summary](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#summary-page) and [Raw](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#raw-page) pages, values from all/selected rows are automatically aggregated in the column header for applicable metrics. Selected individual cells are aggregated in the bottom status bar. - Added `Launch Name` and `Device` options in the filter dialog launched by `Apply Filters` button in the [report header](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-header). - Added support for [source view profiles](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiles) that persist the Source page configuration and allow you to re-apply it to other reports. - The [Metric Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#metric-details) tool window now supports querying metrics beyond the current report by using the `chip:` tag in the search. - Added support for _CUDA Graph Edge Data_ (such as port and dependency type) and _CUDA Graph Conditional Handles_ in the [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#resources) tool window. - The [Acceleration Structure Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#acceleration-structure-viewer) and [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#resources) tool window now support OptiX Opacity Micromaps. **NVIDIA Nsight Compute CLI** - Tracking and profiling all child processes (`--target-processes all`) is now the default for ncu. - Improved reporting of requested but unavailable metrics. Metrics requested in section files are by default considered optional and only cause a warning to be shown. **Resolved Issues** - Support for tracking child processes launched with `system()` is available on Linux ppc64le. - Improved the behavior of following SASS navigation links on the Source page. - Fixed issues with profiling CUDA graphs in graph-profiling mode when nodes are associated with a non-current CUDA context. - Fixed an issue in L2 bandwidth calculations in the hierarchical roofline sections. #### Updates in 2023.2.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-2-2 "Permalink to this headline") **Resolved Issues** - Fixed possible crash when profiling CUDA graphs on multiple GPUs. - Fixed the detection mechanism of the C2C interface, i.e. `metric c2clink__present`. The fix requires the display driver shipping with this release or any newer driver. #### Updates in 2023.2.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-2-1 "Permalink to this headline") **Resolved Issues** - Fixed a crash during application replay while having the temporary directory located on a network file system (NFS). - Improved detection mechanism for C2C interface. Added caching of the detected configuration to reduce overhead. #### Updates in 2023.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-2 "Permalink to this headline") **General** - Extended the rules system to show estimates of the potential speedup that can be achieved by addressing the corresponding performance bottleneck. These speedups allow prioritizing applicable rules and help focusing first on optimization strategies with the highest potential performance gain. - Added support for rules to highlight individual source lines. Lines with global/local memory access with high excessive sector counts and shared accesses with many bank conflicts are automatically detected and highlighted. - Added the ability to query metric attributes in NvRules API. - Added support for creating instanced metrics through the NvRules API. - For Orin+ mobile chips on the Linux aarch64 platform, added metrics (`mcc__*`) support for memory controller channel (MC Channel) unit which connects to the DRAM. **NVIDIA Nsight Compute** - Added hyperlinks to the SASS View of the Source Page for instructions that reference others by address or offset. This enables to quickly jump to the target instruction of a branch. - Improved the search bar in the Metric Details tool window. The search string now matches any part of the metric names, and the matching results are shown in a sorted order. - Added a visual indication of scale of the metric value changes when the baselines are used. The background bars in the table cells of the Details Page allow to quickly identify which metrics values increased or decreased the most. The color scheme can be configured in the [Baselines tool window](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-baselines). - Added support for rules toggle button on the Summary Page. Allows to hide the bottom pane with the rules output for the selected kernel launch. - Added support for allowing users to configure properties on [Summary Page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-summary-page) using [Metrics/Properties profile option](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#options-profile). - Added percentage bars on [Summary Page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-summary-page). **NVIDIA Nsight Compute CLI** - Added support for tracking child processes launched with `posix_spawn(p)` when using `--target-processes all`. - Added support for tracking child processes launched with `system()` on Windows and Linux (aarch64, x86\_64) when using `--target-processes all`. **Resolved Issues** - Fixed table alignment in the output of the NVIDIA Nsight Compute CLI on Windows when printing Unicode characters. - Fixed view corruption in the Source Page after switching from the collapsed view to the expanded view. - Fixed missing tooltip descriptions for some SASS instructions. - Fixed potential crash when copying from the Resources tool window using CTRL+C. - Fixed possible crash when restoring sections in the Sections tool window. #### Updates in 2023.1.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-1-1 "Permalink to this headline") **NVIDIA Nsight Compute** - Added new configuration options to set the default view mode and precision for the Source page. **Resolved Issues** - Added support for the `DT_RUNPATH` attribute when intercepting calls to `dlopen`. Fixed issue for applications or libraries relying on `DT_RUNPATH` not finding all dynamic libraries when launched by NVIDIA Nsight Compute. - Improved interaction between custom additional metrics and the selected metric set. Adding custom metrics no longer forces switching to the custom metric set. - Added ability to gracefully skip folders with insufficient access permissions while importing source code. - Fixed the calculation of the peak values for the L1 and L2 cache bandwidths in the hierarchical roofline charts. - Fixed issue that prevented modules loaded with function `optixModuleCreateFromPTX` showing up in the _Optix: Modules_ table of the _Resources_ tool window. - Fixed handling of deprecated functions when querying function pointers from the OptiX interception library. - Fixed that sometimes sections or rules couldn’t be easily selected in the tool window. - Fixed issue with _Reset Application Data_ that prevented some setting from correctly resetting. - Fixed potential crash of NVIDIA Nsight Compute when _Reset Application Data_ was executed multiple times in a row. - Fixed a crash when saving or loading baselines for non-kernel results. - Fixed that memory written while executing a CUDA graph was not properly restored in single-pass graph profiling. - Fixed potential memory leak while collecting SW counters for modules with unpatched kernel functions. #### Updates in 2023.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2023-1 "Permalink to this headline") **General** - Added support for the CUDA toolkit 12.1. - Added a new [app-range](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#application-range-replay) replay mode to profile ranges without API capture by relaunching the entire application multiple times. - Added _sharedBankConflicts_ sample CUDA application and document to show how NVIDIA Nsight Compute can be used to analyze and identify the shared memory bank conflicts which result in inefficient shared memory accesses. Refer to the `README.TXT` file, sample code and document under `extras/samples/sharedBankConflicts`. - Jupyter notebook samples are available in the Nsight training [github repository](https://github.com/NVIDIA/nsight-training/blob/master/cuda/nsight_compute/python_report_interface). - The equivalent of the [high-level Python report interface](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#python-report-interface-high-level) is now available in rule files. **NVIDIA Nsight Compute** - Added support for profiling individual metrics in [Interactive Profile activity](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-activity-interactive). A new input field for metrics was added in the [Metric Selection](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-sections-info) tool window. - Files on remote systems can be opened directly from the [menu](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#main-menu). - Metric- and section-related entries in the menu, [Profile activity](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-activity-non-interactive) and [Metric Selection](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-sections-info) tool window were renamed to make them more clear. - CPU and GPU [NUMA topology metrics](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#metrics-reference) can be collected on applicable systems. Topology information is shown in a new [NUMA Affinity section](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#sections-and-rules). - Added content-aware suggestions to the Details page to provide suggestions based on the selected profiling options. - Added support for [re-resolving source files](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-navigation) on the Source page. - Not-issued warp stall reasons are removed from the Source Counters section tables and hidden by default on the Source page. Users should focus on regular warp stall reasons by default and only inspect not-issued samples if this distinction is needed. - Added support to search missing CUDA source files to permanently import into the report using [Source Lookup options](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#options-source-lookup) in the [Interactive Profile activity](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-activity-interactive). - The [source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page-metrics) now shows metric values as percentages by default. New buttons are added to support switching between different value modes. **NVIDIA Nsight Compute CLI** - Added support for config files in the current working or user directory to set default ncu parameters. See the [General options](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-general) for more details. - Added `--range-filter` [command line option](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-console-output) which allows to select subset of enabled profile ranges. - Added new `--source-folders` [command line option](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-profile) that allows to recursively search for missing CUDA source files to permanently import into the report. **Resolved Issues** - Fixed performance issues on the Summary and Raw pages for large reports. - Improved support for non-ASCII characters in filenames. - Fixed an issue with delayed updates of assembly analysis information on the Source page’s Source and PTX views. - Fixed potential crashes when using the Python report interface. #### Updates in 2022.4.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-4-1 "Permalink to this headline") **General** - Improved the documentation for the NvRules API. - The python report interface links libstdc++ statically. **Resolved Issues** - Fixed an issue that enabled profiling on CUDA Graph uploads. - Fixed formatting issues during unit conversion of metric instances. - Fixed an issue that could lead to a crash during application replay. - Fixed an issue that could lead to a crash in the python report interface. - Fixed typos in the metrics reference documentation and descriptions. #### Updates in 2022.4 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-4 "Permalink to this headline") **General** - Added support for the CUDA toolkit 12.0. - Added support for profiling [CUDA graphs](https://docs.nvidia.com/cuda/cuda-driver-api/group__CUDA__GRAPH.html.md#group__CUDA__GRAPH) as complete workloads instead of as single kernel nodes. Enable this using the _Graph Profiling_ option in the [activities](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-dialog). Similarly to [range replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay) results, selected metrics are not available when profiling graphs. - Added support for profiling on Windows Subsystem for Linux (WSL2). See the [System Requirements](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html.md#system-requirements) for more details. - Deprecated `nv-nsight-cu` and `nv-nsight-cu-cli` aliases are removed in favor of `ncu-ui` and `ncu`. **NVIDIA Nsight Compute** - The [Source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) now loads disassembly and static analysis results asynchronously in the background. - Added a new [Metric Details](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-metric-details) tool window to inspect metric information such as raw value, unit, description or instance values. Open the tool window and select a metric on the _Details_ or _Raw_ page or lookup any metric in the focused report directly in the tool window’s search bar. - In the Source page PTX view, the source name will be shown as a list of comma-separated files. - Added flexibility with NVTX based filtering in the _Next Trigger_ filter, similar to the command line. Filters can now use nvtx-include and nvtx-exclude expressions by adding the `nvtx-include:` or `nvtx-exclude:` prefix. - NVTX views now show the payload type. - Simplified the command line generated by the [Profile activity](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-activity-non-interactive). - Reduced the number of steps required to re-run the [Profile activity](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#connection-activity-non-interactive). - The way to rename [Baselines](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#baselines) in-place has been improved. - The [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window now shows the CUDA Dynamic Parallelism state for CUDA functions and modules. - OptiX traversable handles can now be exported as [Graphviz](https://graphviz.org/) DOT or SVG files for visualization from the Resources tool window. - All OptiX build, instance and geometry flags can be viewed in the [Acceleration Structure Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#acceleration-structure-viewer). - Added OptiX-specific highlight filters to the Acceleration Structure Viewer. - Added support for user-specified index strides to the Acceleration Structure Viewer. **NVIDIA Nsight Compute CLI** - Added new option `--graph-profiling graph` to enable profiling of complete CUDA graphs as single workloads. - Added new option `--filter-mode per-launch-config` to enable filtering of kernel launches for each GPU launch parameter separately. - Added support to print section body item metrics on the details page with the new `--print-details` [command line option](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-console-output). - Added support to select what to show in Metric Name column on the details page with the new `--print-metric-name` [command line option](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-console-output). - Removed deprecated options: `--units`, `--fp`, `--summary` and `--kernel-base` - Added support to print launch, session, process and device attributes on session page with the new `--page session` option. - Added `--kill yes` support for application replay mode. **Resolved Issues** - Fixed an issue that NVIDIA Nsight Compute could crash when continuing profiling after transposing the _Raw_ page table. - Fixed an issue that caused closing a report document to be delayed by pending source analysis. - Fixed support for profiling applications with older OptiX versions. - Fixed display of OptiX module inputs for IR and built-in modules. #### Updates in 2022.3 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-3 "Permalink to this headline") **General** - Added support for the CUDA toolkit 11.8. - Added support for the Ada GPU architecture. - Added support for the Hopper GPU architecture. - Added support for [OptiX 7.6](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html.md#library-support-optix). - Added _uncoalescedGlobalAccesses_ sample CUDA application and document to show how the NVIDIA Nsight Compute profiler can be used to analyze and identify the memory accesses which are uncoalesced and result in inefficient DRAM accesses. Refer to the README, sample code and document under `extras/samples/uncoalescedGlobalAccesses`. - Added [Metrics Reference](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#metrics-reference) in the documentation that lists metrics not available through `--query-metrics`. - Reduced the overhead of collecting SASS-patching based metrics. - On [Multi-Instance GPU (MIG)](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#special-configurations-mig) configurations, NVIDIA Nsight Compute cannot lock clocks anymore. Users are expected to lock clocks externally using nvidia-smi. **NVIDIA Nsight Compute** - Wrapper script `nv-nsight-cu` is deprecated in favor of `ncu-ui` and will be removed in a future release. - Source page supports range replay results. - Added a second chart on the Compute Workload Analysis section to avoid mixing metrics with different meaning. - NVIDIA Nsight Compute now tracks traversable handles created with `optixAccelRelocate`. - NVIDIA Nsight Compute now tracks traversable handles created as updates from others. - The Acceleration Structure viewer now reports unsupported inputs. - The Acceleration Structure viewer now supports opening multiple traversable handles. - The Acceleration Structure viewer now uses OptiX naming for displayed elements. **NVIDIA Nsight Compute CLI** - Wrapper script `nv-nsight-cu-cli` is deprecated in favor of `ncu` and will be removed in a future release. - Added new option `--filter-mode per-gpu` to enable filtering of kernel launches on each GPU separately. - Added new option `--app-replay-mode relaxed` to produce profiling results for valid kernels even if the number of kernel launches is inconsistent across application replay passes. - Added a documentation section on supported [environment variables](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#environment-variables). - Improved the performance when loading existing reports on the command line. **Resolved Issues** - Fixed an issue when resolving files on the Source page. - Fixed an issue when profiling OptiX applications. - Fixed an issue in the OptiX traversable handle management caused by clashing handle values. - Fixed an issue in the Acceleration Structure viewer causing the display of invalid memory when viewing AABB buffers. #### Updates in 2022.2.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-2-1 "Permalink to this headline") **Resolved Issues** - Fixed an issue that caused some tootips to not show up for the charts on the Details page. - Fixed the incorrect reporting of the accessed bytes for LDGSTS (access) traffic in the L1TEX memory table. - Fixed an issue that resulted in an empty view on the Source page after resolving multiple source files at once. - Fixed a failure to connect to remote machines over SSH due to a mismatch in the configuration of data compression. - Fixed a potential failure to profile kernels on multiple devices of the same type of chip. The failure occurred on the attempt to profile on the second device. #### Updates in 2022.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-2 "Permalink to this headline") **General** - Added support for the CUDA toolkit 11.7. - Improved performance for profiling and metric query. - Added Linux (aarch64 sbsa) as a supported [host platform](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html.md#platform-support). - The NVIDIA Nsight Compute CLI stores the command line arguments, which can be viewed in the [Session](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-session-page) report page. - Added an API to query the version of the [Python Report](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#python-report-interface) and [NvRules](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md#abstract) interfaces. - Added an API to query the PTX in the [Python Report](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#python-report-interface) and [NvRules](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md#abstract) interfaces. **NVIDIA Nsight Compute** - The [Acceleration Structure Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#acceleration-structure-viewer) allows inspection of acceleration structures built using the OptiX API for debugging and performance optimization. - The [Source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) column chooser now supports to enable or disable groups of metrics. Note that not all metrics are enabled anymore by default to make the view easier to use. - The [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window now links to the exact target resource instances for _CUDA_ resource types. - The [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window now shows the instanced nodes for CUDA graphs. - The [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window now shows the loading state and number of loaded functions for _CUDA Modules_. - The [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window now shows the graph node enablement state for applicable instanced graph nodes. - The [Resources](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-resources) tool window now shows the graph node priorities for instanced kernel graph nodes. - Added regex support in the _Next Trigger_ filter for NVTX based filtering. The _Next Trigger_ filter now considers the NVTX config as a regular expression if the `regex:` prefix is specified. - Added regex support in the report’s _Filter Results_ dialog. - Added [keyboard shortcuts](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#quick-start-navigate-report) to navigate between the pages in a report. - The behavior for selecting sets and sections is now consistent between the [Sections/Rules Info window](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-sections-info) and the [non-interactive profile activity](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#quick-start-non-interactive). - Reports can now be opened directly from the welcome dialog. **NVIDIA Nsight Compute CLI** - Added support for collecting [sampling-based](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#sampling) warp stalls in range replay mode. - Added regex support in [NVTX filtering](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#nvtx-filtering). - The metric type is shown when querying metrics. **Resolved Issues** - Reduced overhead of connecting to the host UI for non-interactive remote profiling sessions. - Fixed issues with persisting the Source page state when collapsing or switching between results. - Fixed an issue that locked GPU clocks were not reset when terminating the NVIDIA Nsight Compute CLI while profiling a kernel. - Fixed issues with selecting and copying text from the Details page tables. - Fixed an issue with opening report files in the UI on macOS. - Fixed an issue with the _Freeze API_ option. #### Updates in 2022.1.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-1-1 "Permalink to this headline") **General** - Filtering kernel launches or profile results based on NVTX domains/ranges now takes registered strings in the payload field into account, if the range name is empty. - Added support for the suffix `.max_rate` for ratio metrics. **Resolved Issues** - Fixed a crash during the disassembly of the kernel’s SASS code for the Source page. - Fixed a crash on exit of the NVIDIA Nsight Compute UI. - Fixed a hang during profiling when CPU call stack collection is enabled. - Fixed missing to flush UVM buffers before taking memory checkpoints during [Range Replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay). - Fixed tracking of memory during [Range Replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay), if the CUDA context has any device mapped memory allocations. - Fixed the maximum available shared memory sizes in the [Occupancy Calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#occupancy-calculator) for NVIDIA Ampere GPUs. - Fixed that the shared memory usage of the kernel is incorrectly initialized when opening the [Occupancy Calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#occupancy-calculator) from a profile report. #### Updates in 2022.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2022-1 "Permalink to this headline") **General** - Added support for the CUDA toolkit 11.6. - Added support for GA103 chips. - Added a new [Range Replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#range-replay) mode to profile ranges of multiple, concurrent kernels. Range replay is available in the NVIDIA Nsight Compute CLI and the non-interactive Profile activity. - Added a new rule to detect non-fused floating-point instructions. - The Uncoalesced Memory access rules now show results in a dynamic table. - Unix Domain Sockets and Windows Named Pipes are used for local connection between the host and target processes on x86\_64 Linux and Windows, respectively. - The [NvRules API](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md#abstract) now supports querying action names using different function name bases (e.g. demangled). **NVIDIA Nsight Compute** - The default [report page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-pages) is now chosen automatically when opening a report. - Added coverage for ECC (Error Correction Code) operations in the L2 Cache table of the Memory Analysis section. - Added a new [L2 Evict Policies](https://docs.nvidia.com/nsight-compute/ProfilingGuide/topics/memory-tables-l2-evict-policy.html.md) table to the Memory Analysis section. - The [Occupancy Calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#occupancy-calculator) now updates automatically when the input changes. - Added new metric _Thread Instructions Executed_ to the Source page. - Added tooltips to the [Register Dependency](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) columns in the Source page to identify the associated register more conveniently. - Improved the selection of Sections and Sets in the Profile activity connection dialog. - NVLink utilization is shown in the NVLink Tables section. - NVLink links are colored according to the measured throughput. **NVIDIA Nsight Compute CLI** - `--kernel-regex` and `--kernel-regex-base` options are no longer supported. Alternate options are `--kernel-name` and `--kernel-name-base` respectively, added in 2021.1.0. - Added support to resolve CUDA source files in the `--page source` output with the new `--resolve-source-file` [command line option](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-console-output). - Added new option `--target-processes-filter` to filter the processes being profiled by name. - The CPU Stack Trace is shown in the NVIDIA Nsight Compute CLI output. **Resolved Issues** - Fixed the calculation of aggregated average instruction execution metrics in non-SASS views on the Source page. - Fixed that atomic instructions are counted as both loads and stores in the Memory Analysis tables. #### Updates in 2021.3.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-3-1 "Permalink to this headline") **Resolved Issues** - Fixed that kernels with the same name and launch configuration were in some scenarios associated with the wrong profiling results during application replay. - Fixed an issue with binary forward compatibility of the report format. - Fixed an issue with applications calling into the CUDA API during process teardown. - Fixed an issue profiling application using pre-CUDA API 3.1 contexts. - Fixed a crash when resolving files on the Source page. - Fixed that opening reports with large embedded CUBINs would hang the UI. - Fixed an issue with remote profiling on a target where the UI is already launched. #### Updates in 2021.3 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-3 "Permalink to this headline") **General** - Added support for the CUDA toolkit 11.5. - Added a new rule for detecting inefficient memory access patterns in the L1TEX cache and L2 cache. - Added a new rule for detecting high usage of system or peer memory. - Added new `IAction::sass_by_pc` function to the [NvRules API](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md#abstract). - The [Python-based report interface](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#python-report-interface) is now available for Windows and macOS hosts, too. - Added Hierarchical Roofline section files in a new “roofline” section set. - Added support for collecting CPU call stack information. **NVIDIA Nsight Compute** - Added support for new remote profiling [SSH connection and authentication options](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#remote-connections) as well as local SSH configuration files. - Added an [Occupancy Calculator](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#occupancy-calculator) which can be opened directly from a profile report or as a new activity. It offers feature parity to the CUDA Occupancy Calculator [spreadsheet](http://docs.nvidia.com/cuda/cuda-occupancy-calculator/index.html). - Added new [Baselines tool window](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#tool-window-baselines) to manage (hide, update, re-order, save/load) baseline selections. - The Source page views now support multi-line/cell selection and copy/paste. Different colors are used for highlighting selections and correlated lines. - The search edit on the Source page now supports _Shift+Enter_ to search in reverse direction. - The [Memory Workload Analysis Chart](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#memory-chart) can be configured to show throughput values instead of transferred bytes. - The _Profile_ activity now supports the `--devices` option. - The _NVLink Topology_ diagram displays per NVLink metrics. - Added a new tool window showing the CPU call stack at the location where the current thread was suspended during interactive profiling activities. - If enabled, the _Call Stack / NVTX_ page of the profile report shows the captured CPU call stack for the selected kernel launch. **NVIDIA Nsight Compute CLI** - Added support for printing source/metric content with the new `--page source` and `--print-source` [command line options](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#command-line-options-console-output). - Added new option `--call-stack` to enable collecting the CPU call stack for every profiled kernel launch. **Resolved Issues** - Fixed that `memory_*` metrics could not be collected with the `--metrics` option. - Fixed that selection and copy/paste was not supported for section header tables on the Details page. - Fixed issues with the Source page when collapsing the content. - Fixed that the UI could crash when applying rules to a new profile result. - Fixed that PC Sampling metrics were not available for _Profile Series_. - Fixed that local profiling did not work if no non-loopback address was configured for the system. - Fixed termination of remote-launched applications. On QNX, terminating an application profiled via _Remote Launch_ is now supported. Canceling remote-launched _Profile_ activities is now supported. #### Updates in 2021.2.8 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-8 "Permalink to this headline") **General** - Updated Python libraries to version 3.10.5. #### Updates in 2021.2.7 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-7 "Permalink to this headline") **General** - Enabled stack canaries with random canary values for L4T builds. #### Updates in 2021.2.6 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-6 "Permalink to this headline") **Resolved Issues** - Fixed an issue causing a hang on QNX after pressing `ctrl+c` while profiling a multi-process application. #### Updates in 2021.2.5 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-5 "Permalink to this headline") **Resolved Issues** - Improve the handling of the performance monitor reservation on mobile target GPUs. #### Updates in 2021.2.4 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-4 "Permalink to this headline") **Resolved Issues** - Fixed an issue that prevented remote interactive profiling of kernels on NVIDIA GA10b chips. #### Updates in 2021.2.3 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-3 "Permalink to this headline") **General** - Added support for the NVIDIA GA10b chip. **Resolved Issues** - Improved error message on QNX for failure to deploy stock section and rules files. #### Updates in 2021.2.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-2 "Permalink to this headline") **General** - Changes for profiling support on NVIDIA virtual GPUs (vGPUs) for an upcoming GRID/vGPU release. **Resolved Issues** - Fixed hang issue on QNX when using the `--target-processes all` option while profiling shell scripts. #### Updates in 2021.2.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2-1 "Permalink to this headline") **General** - Reduced the memory overhead when loading reports in the [Python Report Interface](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#python-report-interface). **Resolved Issues** - Fixed that links in the _Memory Allocations_ Resource view were not working correctly. - Fixed that NVTX state might not be correctly reset between interactive profiling activities. - Fixed that the UI could crash when opening baselines from different GPU architectures. #### Updates in 2021.2 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-2 "Permalink to this headline") **General** - Added support for the CUDA toolkit 11.4. - Added support for OptiX version 7.3. - Added support for profiling on [NVIDIA virtual GPUs](https://www.nvidia.com/en-us/data-center/virtual-gpu-technology.md/) (vGPUs) on an upcoming GRID/vGPU release. - Added a new [Python-based report interface](https://docs.nvidia.com/nsight-compute/CustomizationGuide/index.html.md#python-report-interface) for interacting with report files from Python scripts. - Added a new rule to warn users when sampling metrics were selected, but no sampling data was collected. - Renamed _SOL_ to _Throughput_ in the Speed of Light section. - Renamed several `memory_*` metrics used on the _Source_ page, to better reflect the measured value. See the [Source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page) documentation for more details. **NVIDIA Nsight Compute** - Added support for opening [cubin files](https://docs.nvidia.com/cuda/cuda-binary-utilities/index.html.md#cuda-binary) in a [Standalone Source Viewer](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#cubin-viewer) without profiling the application. - Moved the output of all rules so that it is visible even if a section’s body is collapsed. Visibility of the rules’ output can be toggled by a new button in the report header. - The profiler report header now shows the report name for each baseline when ambiguous. - Rules can define _Focused Metrics_ that were most important for triggering their result output. Metrics are provided per result message which additional information, such as the underlying conditions and thresholds. - [Memory tables](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#memory-tables) show tooltips for cells with derived metric calculations. - Added a knowledge base service to show more comprehensive background information on metric names and descriptions in their tooltips. - Following a link in the Source Counters hot spot tables automatically selects the corresponding metric in the Source page. - Added new columns for visualizing register dependencies in the SASS view of the [Source page](https://docs.nvidia.com/nsight-compute/NsightCompute/index.html.md#profiler-report-source-page). - Functions in the SASS view are now sorted by name. - Added support for OptiX 7.x resource tracking in the interactive profile activity. The _Resources_ tool window will show information on instantiated `optixDeviceContexts, optixModules, optixProgramGroups, optixPipelines and optixDenoiser` objects. - Added support for new CUDA graph memory allocation APIs. - Improved consistency between command line parameters and the _Next Trigger_ filter in the API Stream window for handling of regex inputs. The _Next Trigger_ filter now considers kernel/API name as a regular expression only if string has `regex:` as prefix. - Added ability to select font settings in the options dialog. - Added ability to configure the metrics shown on the summary page via the options dialog. - The selected heatmap color scale now also applies to the _Memory chart_. - The ncu-ui script now checks for missing library dependencies, such as OpenGL or [Qt](https://doc.qt.io/qt-5/linux-requirements.html). **NVIDIA Nsight Compute CLI** - Added environment variable [NV\_COMPUTE\_PROFILER\_DISABLE\_STOCK\_FILE\_DEPLOYMENT=1](https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html.md#environment-variables) to skip deployment of section and rule files. **Resolved Issues** - Fixed a performance issue in the NVIDIA Nsight Compute CLI when using `--page raw --csv --units auto`. - Fixed that the SSH passphase key is no longer persisted in the project file. - Fixed state of restore button in connection dialog. The button now supports restoring the default settings, if current setting differ from the default. - Fixed that the complete GPU name can be shown in the NVLINK topology diagram on macOS. - Fixed that collapsing the Source view reset the selected metrics. - Fixed that correlated lines could differ between filtered and unfiltered views of the executed functions. - Fixed that two application icons were shown in the macOS dock. - Improved HiDPI awareness. #### Updates in 2021.1.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-1-1 "Permalink to this headline") **General** - Updated OpenSSL library to version 1.1.1k. **NVIDIA Nsight Compute** - Remote source resolution can now use the IP address, in addition to the hostname, to find the necessary SSH target. **NVIDIA Nsight Compute CLI** - Added support for the existing command line options for kernel filtering while importing data from an existing report file using `--import`. - Option `-k` is not considered as deprecated option `--kernel-regex` anymore. **Resolved Issues** - Fixed failure to profile kernels from applications that use the CUDA graphics interop APIs to share semaphores. - Fixed wavefront metric in the L1TEX table for writes to shared memory on GA10x chips. - Fixed an issue resulting in incomplete data collection for the interactive profile activity after switching from single-pass mode to collecting multiple passes in the same session. - Fixed values shown in the mimimap of the Source page when all functions are collapsed. - Fixed an issue causing names set by the NVTX naming APIs of one application to be applied to all subsequent sessions of the same instance of NVIDIA Nsight Compute. - Fixed behavior of horizontal scroll bars when clicking in the source views on the Source page. - Fixed appearance of multi-line entries in column chooser on the Source page. - Fixed enablement state of the reset button on the Connection dialog. - Fixed potential crash of NVIDIA Nsight Compute when windows size becomes small while being on the Source page. - Fixed potential crash of NVIDIA Nsight Compute when relative paths for section/rules files could not be found. - Fixed potential crash of NVIDIA Nsight Compute after removing baselines. #### Updates in 2021.1 [](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html\#updates-in-2021-1 "Permalink to this headline") **General** - Added support for the CUDA toolkit 11.3. - Added support for the [OptiX 7 API](https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html.md#library-support-optix). - `GpuArch` enumeration values used for filtering in section files were renamed from architecture names to compute capabilities. - NVTX states can now be accessed via the [NvRules API](https://docs.nvidia.com/nsight-compute/NvRulesAPI/index.html.md#abstract). - Added a rule for the _Occupancy_ section. **NVIDIA Nsight Compute** - Added support for new CUDA asynchronous allocator attributes in the _Memory Pools_ resources view. - Added a topology chart and link properties table in the NVLink section. - The selected metric column is scrolled into view on the _Source_ page when a new metric is selected. - Users can choose the _Source_ heatmap color scale in the _Options_ dialog. **NVIDIA Nsight Compute CLI** - Added file-based [application replay](https://docs.nvidia.com/nsight-compute/ProfilingGuide/index.html.md#application-replay) as the new default application replay mode. File-based replay uses a temporary file for keeping replay data, instead of allocating them in memory. This keeps the required memory footprint close to constant, independent of the number of profiled kernels. Users can switch between buffer modes using the `--app-replay-buffer` option. - CLI output now shows NVTX color and message information. - `--kernel-regex` and `--kernel-regex-base>` options are deprecated and replaced by `--kernel-name` and `--kernel-regex-base`, respectively. - All options which support regex need to provide `regex:` as a prefix before an argument to match per the regex, e.g `