User Guide
NVIDIA Nsight Systems user guide
Profiling from the CLI
Installing the CLI on Your Target
The Nsight Systems CLI provides a simple interface to collect on a target without using the GUI. The collected data can then be copied to any system and analyzed later.
The CLI is distributed in the Target directory of the standard Nsight Systems download package.
If you wish to run the CLI without root (recommended mode), you will want to install in a directory where you have full access.
Note that you must run the CLI on Windows as administrator.
Command Line Options
The Nsight Systems command lines can have one of two forms:
nsys [global_option]
or
nsys [command_switch][optional command_switch_options][application] [optional application_options]
All command line options are case sensitive. For command switch options, when short options are used, the parameters should follow the switch after a space; e.g. -s process-tree. When long options are used, the switch should be followed by an equal sign and then the parameter(s); e.g. --sample=process-tree.
For this version of Nsight Systems, if you launch a process from the command line to begin analysis, the launched process will be terminated when collection is complete, including runs with --duration set, unless the user specifies the --kill none option (details below). The exception is that if the user uses NVTX, cudaProfilerStart/Stop, or hotkeys to control the duration, the application will continue unless --kill is set.
The Nsight Systems CLI supports concurrent analysis by using sessions. Each Nsight Systems session is defined by a sequence of CLI commands that define one or more collections (e.g. when and what data is collected). A session begins with either a start, launch, or profile command. A session ends with a shutdown command, when a profile command terminates, or, if requested, when all the process tree(s) launched in the session exit. Multiple sessions can run concurrently on the same system.
CLI Global Options
Short |
Long |
Description |
|---|---|---|
-h |
|
Help message providing information about available command switches and their options. |
-v |
|
Output Nsight Systems CLI version information. |
CLI Command Switches
The Nsight Systems command line interface can be used in two modes. You may
launch your application and begin analysis with options specified to the
nsys profile command. Alternatively, you can control the launch of an
application and data collection using interactive CLI commands.
Command |
Description |
|---|---|
analyze |
Post process existing Nsight Systems result, either in .nsys-rep or SQLite format, to generate expert systems report. |
cancel |
Cancels an existing collection started in interactive mode. All data already collected in the current collection is discarded. |
export |
Generates an export file from an existing .nsys-rep file. For more information about the exported formats see the /documentation/nsys-exporter directory in your Nsight Systems installation directory. |
launch |
In interactive mode, launches an application in an environment that supports the requested options. The launch command can be executed before or after a start command. |
nvprof |
Special option to help with transition from legacy NVIDIA nvprof tool. Calling |
profile |
A fully formed profiling description requiring and accepting no further input. The command switch options used (see below table) determine when the collection starts, stops, what collectors are used (e.g. API trace, IP sampling, etc.), what processes are monitored, etc. |
recipe |
Post process multiple existing Nsight Systems results to generate statistical information and create various plots. See Multi-Report Analysis topic for details. |
sessions |
Gives information about all sessions running on the system. |
shutdown |
Disconnects the CLI process from the launched application and forces the CLI process to exit. If a collection is pending or active, it is cancelled |
start |
Start a collection in interactive mode. The start command can be executed before or after a launch command. |
stats |
Post process existing Nsight Systems result, either in .nsys-rep or SQLite format, to generate statistical information. |
status |
Reports on the status of a CLI-based collection or the suitability of the profiling environment. |
stop |
Stop a collection that was started in interactive mode. When executed, all active collections stop, the CLI process terminates but the application continues running. |
CLI Profile Command Switch Options
After choosing the profile command switch, the following options are available. Usage:
nsys [global-options] profile [options] [application] [application-arguments]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
none,tegra-accelerators |
none |
Collect other accelerators workload trace from the hardware engine units. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
true, false |
false |
Derive report file name from collected data uses details of profiled graphics application. Format: [Process Name][GPU Name][Window Resolution][Graphics API] Timestamp .nsys-rep If true, automatically generate report file names. |
|
-b |
|
auto,fp,lbr,dwarf,none |
Select the backtrace method to use while sampling. The option ‘lbr’ uses Intel(c) Corporation’s Last Branch Record registers, available only with Intel(c) CPUs codenamed Haswell and later. The option ‘fp’ is frame pointer and assumes that frame pointers were enabled during compilation. The option ‘dwarf’ uses DWARF’s CFI (Call Frame Information). Setting the value to ‘none’ can reduce collection overhead. |
|
-c |
|
none, cudaProfilerApi, hotkey, nvtx |
none |
When |
|
none, stop, stop-shutdown, repeat[:N], repeat-shutdown:N |
stop-shutdown |
Specify the desired behavior when a capture range ends. Applicable only when used along with
|
|
|
true, false |
false |
Collect clock frequency changes. Available only in Nsight Systems Embedded Platforms Edition and Arm server (SBSA) platforms |
|
|
< filename > |
none |
Open a file that contains profile switches and parse the switches. Note additional switches on the command line will override switches in the file. This flag can be specified more than once. |
|
|
0x16, 0x17, …, none |
none |
Collect per-cluster Uncore PMU counters. Multiple values can be selected, separated by commas only (no
spaces). Use the |
|
|
0x11,0x13,…,none |
none |
Collect per-core PMU counters. Multiple values can be selected, separated by commas only (no spaces). Use
the |
|
|
‘help’ or the end users selected events in the format ‘x,y’ |
‘2’ i.e. Instructions Retired |
Select the CPU Core events to sample. Use the |
|
|
0,1,2,…,none |
none |
Collect metrics on the CPU core. Multiple values can be selected, separated by commas only (no spaces). Use
the |
|
|
0x2a, 0x2c, …, none |
none |
Collect per-socket Uncore PMU counters. Multiple values can be selected, separated by commas only (no spaces).
Use the |
|
|
process-tree, system-wide, none |
process-tree |
Trace OS thread scheduling activity. Select ‘none’ to disable tracing CPU context switches. Depending on
the platform, some values may require admin or root privileges. Note: if the |
|
|
milliseconds |
See Description |
Set the interval, in milliseconds, when buffered CUDA data is automatically saved to storage. CUDA data buffer saves may cause profiler overhead. Buffer save behavior can be controlled with this switch. If the CUDA flush interval is set to 0 on systems running CUDA 11.0 or newer, buffers are saved when they fill. If a flush interval is set to a non-zero value on such systems, buffers are saved only when the flush interval expires. If a flush interval is set and the profiler runs out of available buffers before the flush interval expires, additional buffers will be allocated as needed. In this case, setting a flush interval can reduce buffer save overhead but increase memory use by the profiler. If the flush interval is set to 0 on systems running older versions of CUDA, buffers are saved at the end of the collection. If the profiler runs out of available buffers, additional buffers are allocated as needed. If a flush interval is set to a non-zero value on such systems, buffers are saved when the flush interval expires. A cuCtxSynchronize call may be inserted into the workflow before the buffers are saved which will cause application overhead. In this case, setting a flush interval can reduce memory use by the profiler but may increase save overhead. For collections over 30 seconds an interval of 10 seconds is recommended. Default is 10000 for Nsight Systems Embedded Platforms Edition and 0 otherwise. |
|
|
graph, node |
graph |
If ‘graph’ is selected, CUDA graphs will be traced as a whole and node activities will not be collected. This will reduce overhead to a minimum, but requires CUDA driver version 515.43 or higher. If ‘node’ is selected, node activities will be collected, but CUDA graphs will not be traced as a whole. This may cause significant runtime overhead. Default is ‘graph’ if available, otherwise default is ‘node’. |
|
|
true, false |
false |
Track the GPU memory usage by CUDA kernels. Applicable only when CUDA tracing is enabled. Note: This feature may cause significant runtime overhead. |
|
|
true, false |
false |
This switch tracks the page faults that occur when CPU code tries to access a memory page that resides on the device. Note that this feature may cause significant runtime overhead. Not available on Nsight Systems Embedded Platforms Edition. |
|
|
true, false |
false |
This switch tracks the page faults that occur when GPU code tries to access a memory page that resides on the host. Note that this feature may cause significant runtime overhead. Not available on Nsight Systems Embedded Platforms Edition. |
|
|
all, none, kernel, memory, sync, other |
none |
When tracing CUDA APIs, enable the collection of a backtrace when a CUDA API is invoked. Significant runtime overhead may occur. Values may be combined using ‘,’. Each value except ‘none’ may be appended with a threshold after ‘:’. Threshold is duration, in nanoseconds, that CUDA APIs must execute before backtraces are collected, e.g. ‘kernel:500’. Default value for each threshold is 1000ns (1us). Note: CPU sampling must be enabled. |
|
-y |
|
< seconds > |
0 |
Collection start delay in seconds. |
-d |
|
< seconds > |
NA |
Collection duration in seconds, duration must be greater than zero. The launched process will be terminated
when the specified profiling duration expires unless the user specifies the |
|
60 <= integer |
Stop the recording session after this many frames have been captured. Note when it is selected cannot include any other stop options. If not specified, the default is disabled. |
||
|
true, false |
false |
The Nsight Systems trace initialization involves creating a D3D device and discarding it. Enabling this flag makes a call to DXGIDeclareAdapterRemovalSupport() before device creation. Requires DX11 or DX12 trace to be enabled. |
|
|
true, false, individual, batch, none |
individual |
If individual or true, trace each DX12 workload’s GPU activity individually. If batch, trace DX12 workloads’
GPU activity in ExecuteCommandLists call batches. If none or false, do not trace DX12 workloads’ GPU
activity. Note that this switch is applicable only when |
|
|
true, false |
true |
If true, trace wait calls that block on fences for DX12. Note that this switch is applicable only when
|
|
|
true, false |
false |
Enable EL1 sampling. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
< filepath config.json > |
none |
EL1 sampling config. Available in Nsight Systems Embedded Platforms Edition only. |
|
-e |
|
A=B |
NA |
Set environment variable(s) for the application process to be launched. Environment variables should be defined as A=B. Multiple environment variables can be specified as A=B,C=D. |
|
<plugin_name>[,arg1,arg2,…] |
NA |
Use the specified plugin. The option can be specified multiple times to enable multiple plugins.
Plugin arguments are separated by commas only (no spaces). Commas can be escaped with a backslash ‘\’.
The backslash itself can be escaped by another backslash ‘\\’. To include spaces in an argument,
enclose the argument in double quotes ‘”’. To list all available plugins, use |
|
|
“<name>,<guid>”, or path to JSON file |
none |
Add custom ETW trace provider(s). If you want to specify more attributes than Name and GUID, provide a JSON configuration file as as outlined below. This switch can be used multiple times to add multiple providers. Note: Only available for Windows targets. |
|
|
system-wide, none |
none |
Use the |
|
|
Integers from 1 to 20 Hz |
3 |
The sampling frequency used to collect event counts. Minimum event sampling frequency is 1 Hz. Maximum event sampling frequency is 20 Hz. Not available in Nsight Systems Embedded Platforms Edition. |
|
|
arrow, hdf, json, sqlite, text, none |
none |
Create additional output file(s) based on the data collected. This option can be given more than once. WARNING: If the collection captures a large amount of data, creating the export file may take several minutes to complete. |
|
|
true, false |
true |
If set to true, any call to cudaProfilerStop() will cause the CUDA trace buffers to be flushed. Note that the CUDA trace buffers will be flushed when the collection ends, irrespective of the value of this switch. |
|
-f |
|
true, false |
false |
If true, overwrite all existing result files with same output filename (.nsys-rep, .arrows, .h5, .json, .sqlite, .txt). |
|
Collect ftrace events. Argument should list events to collect as: subsystem1/event1,subsystem2/event2. Requires root. No ftrace events are collected by default. |
|||
|
Skip initial ftrace setup and collect already configured events. Default resets the ftrace configuration. |
|||
|
GPU ID, help, all, none |
none |
Collect GPU Metrics from specified devices. Determine GPU IDs by using |
|
|
integer |
10000 |
Specify GPU Metrics sampling frequency. Minimum supported frequency is 10 (Hz). Maximum supported frequency is 200000 (Hz). |
|
|
alias, file:<file name> |
see description |
Specify metric set for GPU Metrics. The argument must be one of the aliases reported by
|
|
|
help, <id1,id2,…>, all, none |
none |
Analyze video devices. |
|
|
true,false |
false |
Trace GPU context switches. Note that this requires driver r435.17 or later and root permission. |
|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
|
‘F1’ to ‘F12’ |
‘F12’ |
Hotkey to trigger the profiling session. Note that this switch is applicable only when
|
|
|
<NIC names> |
none |
A comma-separated list of NIC names. The NICs which |
|
|
<file paths> |
none |
A comma-separated list of file paths. Paths of an existing ibdiagnet db_csv files, containing networks
information data. Nsight Systems will read the networks information from these files. Don’t use ‘~’ alias
within the path. Note that this option should not be used together with the |
|
|
<directory path> |
none |
Sets the path of a directory into which ibdiagnet network discovery data will be written. Use this option
together with the |
|
|
<IB switch GUIDs> |
none |
A comma-separated list of InfiniBand switch GUIDs. Collect InfiniBand switch congestion events from switches identified by the specified GUIDs. This switch can be used multiple times. System scope. Use the –ib-switch-congestion-nic-device, –ib-switch-congestion-percent, and –ib-switch-congestion-threshold-high switches to further control how congestion events are collected. |
|
|
<NIC name> |
none |
The name of the NIC (HCA) through which InfiniBand switches will be accessed. By default, the first active NIC will be used. One way to find a NIC’s name is via the ibnetdiscover –Hca_list | grep “$(hostname)” command. Example usage: –ib-switch-congestion-nic-device=mlx5_3 |
|
|
1 <= integer <= 100 |
50 |
Percent of InfiniBand switch congestion events to be collected. This option enables reducing the network bandwidth consumed by reporting congestion events. |
|
|
1 <= integer <= 1023 |
75 |
High threshold percentage for InfiniBand switch egress port buffer size. Before a packet leaves an InfiniBand switch, it is stored at an egress port buffer. The buffer’s size is checked and if it exceeds the given threshold percentage, a congestion event is reported. The percentage can be greater than 100. |
|
|
<IB switch GUIDs> |
none |
A comma-separated list of InfiniBand switch GUIDs. Collect metrics from the specified InfiniBand switches. This switch can be used multiple times. System scope. |
|
-n |
|
true, false |
true |
When true, the current environment variables and the tool’s environment variables will be specified for the launched process. When false, only the tool’s environment variables will be specified for the launched process. |
|
true,false |
true |
Use detours for injection. If false, process injection will be performed by windows hooks which allows it to bypass anti-cheat software. |
|
|
true, false |
false |
Trace Interrupt Service Routines (ISRs) and Deferred Procedure Calls (DPCs). Requires administrative privileges. Available only on Windows devices. |
|
|
none, sigkill, sigterm, signal number |
sigterm |
Send signal to the target application’s process group. Can be used with |
|
|
openmpi,mpich |
openmpi |
When using |
|
|
true, false |
false |
Collect metrics from supported NIC/HCA devices. System scope. Not available on Nsight Systems Embedded Platforms Edition. |
|
-p |
|
range@domain, range, range@* |
none |
Specify NVTX range and domain to trigger the profiling session. This option is applicable only when
used along with |
|
default, <domain_names> |
Choose to exclude NVTX events from a comma separated list of domains. ‘default’ excludes NVTX events
without a domain. A domain with this name or commas in a domain name must be escaped with ‘\’.
Note: Only one of |
||
|
default, <domain_names> |
Choose to only include NVTX events from a comma separated list of domains. ‘default’ filters the
NVTX default domain. A domain with this name or commas in a domain name must be escaped with ‘\’.
Note: Only one of |
||
|
<json_file> |
Specify the path to the JSON file containing the requested NVTX annotations. |
||
|
true, false |
true |
If true, trace the OpenGL workloads’ GPU activity. Note that this switch is applicable only when
|
|
|
‘help’ or the end users selected events in the format ‘x,y’ |
Select the OS events to sample. Use the |
||
|
integer |
24 |
Set the depth for the backtraces collected for OS runtime libraries calls. |
|
|
integer |
6144 |
Set the stack dump size, in bytes, to generate backtraces for OS runtime libraries calls. |
|
|
nanoseconds |
80000 |
Set the duration, in nanoseconds, that all OS runtime libraries calls must execute before backtraces are collected. |
|
|
< nanoseconds > |
1000 ns |
Set the duration, in nanoseconds, that Operating System Runtime (osrt) APIs must execute before they are traced. Values significantly less than 1000 may cause significant overhead and result in extremely large result files. |
|
-o |
|
< filename > |
report# |
Set report file name. Any %q{ENV_VAR} pattern in the filename will be substituted with the value of the environment variable. Any %h pattern in the filename will be substituted with the hostname of the system. Any %p pattern in the filename will be substituted with the PID of the target process or the PID of the root process if there is a process tree. Any %% pattern in the filename will be substituted with %. Default is report#.{nsys-rep,sqlite,h5,txt,arrows,json} in the working directory. |
|
main, process-tree, system-wide |
main |
Select which process(es) to trace. Available in Nsight Systems Embedded Platforms Edition only. Nsight Systems Workstation Edition will always trace system-wide in this version of the tool. |
|
|
cuda, none, false |
none |
Collect Python backtrace event when tracing the selected API’s trigger. This option is supported
on Arm server (SBSA) platforms and x86 Linux targets. Note: the selected API tracing must be
enabled. For example, |
|
|
true, false |
false |
Collect Python backtrace sampling events. This option is supported on Arm server (SBSA) platforms, x86 Linux and Windows targets. Note: When profiling Python-only workflows, consider disabling the CPU sampling option to reduce overhead. |
|
|
1 < integers < 2000 |
1000 |
Specify the Python sampling frequency. The minimum supported frequency is 1Hz. The maximum
supported frequency is 2KHz. This option is ignored if the |
|
|
class/event,event, class/event:mode, class:mode,help,none |
none |
Multiple values can be selected, separated by commas only (no spaces). See the
|
|
|
system,process,fast,wide |
system:fast |
Values are separated by a colon (‘:’) only (no spaces). ‘system’ and ‘process’ cannot be specified at the same time. ‘fast’ and ‘wide’ cannot be specified at the same time. Please check the QNX documentation to determine when to select the ‘fast’ or ‘wide’ mode. Specify the default mode for QNX kernel events collection. |
|
|
true,false |
true |
Resolve symbols of captured samples and backtraces. |
|
|
true, false |
false |
Retain ETW files generated by the trace, merge and move the files to the output directory. |
|
|
< username > |
none |
Run the target application as the specified username. If not specified, the target application will be run by the same user as Nsight Systems. Requires root privileges. Available for Linux targets only. |
|
-s |
|
process-tree, system-wide, none |
process-tree |
Select how to collect CPU IP/backtrace samples. If ‘none’ is selected, CPU sampling is disabled.
Depending on the platform, some values may require admin or root privileges. If a target application
is launched, the default is ‘process-tree’, otherwise, the default is ‘none’. Note: ‘system-wide’
is not available on all platforms. Note: If set to ‘none’, CPU context switch data will still be
collected unless the |
|
integer <= 32 |
1 |
The number of CPU IP samples collected for every CPU IP/backtrace sample collected. For example, if set to 4, on the fourth CPU IP sample collected, a backtrace will also be collected. Lower values increase the amount of data collected. Higher values can reduce collection overhead and reduce the number of CPU IP samples dropped. If DWARF backtraces are collected, the default is 4, otherwise the default is 1. This option is not available on Nsight Systems Embedded Platforms Edition or on non-Linux targets. |
|
|
100 < integers < 8000 |
1000 |
Specify the sampling/backtracing frequency. The minimum supported frequency is 100 Hz. The maximum supported frequency is 8000 Hz. This option is supported only on QNX, Linux for Tegra, and Windows targets. |
|
|
integer |
determined dynamically |
The number of CPU Cycle events counted before a CPU instruction pointer (IP) sample is collected.
If configured, backtraces may also be collected. The smaller the sampling period, the higher the
sampling rate. Note that smaller sampling periods will increase overhead and significantly increase
the size of the result file(s). Requires |
|
|
integer |
determined dynamically |
The number of events counted before a CPU instruction pointer (IP) sample is collected. The event used to trigger the collection of a sample is determined dynamically. For example, on Intel based platforms, it will probably be “Reference Cycles” and on AMD platforms, “CPU Cycles”. If configured, backtraces may also be collected. The smaller the sampling period, the higher the sampling rate. Note that smaller sampling periods will increase overhead and significantly increase the size of the result file(s). This option is available only on Linux targets. |
|
|
timer, sched, perf, cuda |
timer,sched |
Specify backtrace collection trigger. Multiple APIs can be selected, separated by commas only (no spaces). Available on Nsight Systems Embedded Platforms Edition targets only. |
|
|
[a-Z][0-9,a-Z,spaces] |
profile-<id>-<application> |
Name the session created by the command. Name must start with an alphabetical character followed by
printable or space characters. Any |
|
-w |
|
true, false |
true |
If true, send target process’s stdout and stderr streams to both the console and stdout/stderr files which are added to the report file. If false, only send target process stdout and stderr streams to the stdout/stderr files which are added to the report file. |
|
true,false |
false |
Collect SoC Metrics. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
integer |
10000 |
Specify SoC Metrics sampling frequency. Minimum supported frequency is ‘100’ (Hz). Maximum supported frequency is ‘1000000’ (Hz). Available in Nsight Systems Embedded Platforms Edition only. |
|
|
alias, file:<file name> |
see description |
Specify metric set for SoC Metrics. The argument must be one of the aliases reported by
|
|
|
1 <= integer |
Start the recording session when the frame index reaches the frame number preceding the start frame index. Note when it is selected cannot include any other start options. If not specified, the default is disabled. |
||
-Y |
|
true, false |
false |
Delays collection indefinitely until the nsys start command is executed for this session. Enabling this option overrides the –delay option. |
|
true, false |
false |
Generate summary statistics after the collection. WARNING: When set to true, an SQLite database will be created after the collection. If the collection captures a large amount of data, creating the database file may take several minutes to complete. |
|
-x |
|
true, false |
true |
If true, stop collecting automatically when the launched process has exited or when the duration expires - whichever occurs first. If false, duration must be set and the collection stops only when the duration expires. Nsight Systems does not officially support runs longer than 5 minutes. |
-t |
|
cuda, nvtx, cublas, cublas-verbose, cusparse, cusparse-verbose, cudnn, cudla, cudla-verbose, cusolver, cusolver-verbose, opengl, opengl-annotations, openacc, openmp, osrt, mpi, nvvideo, vulkan, vulkan-annotations, dx11, dx11-annotations, dx12, dx12-annotations, openxr, openxr-annotations, oshmem, ucx, wddm, tegra-accelerators, python-gil, syscall, none |
cuda, opengl, nvtx, osrt |
Select the API(s) to be traced. The osrt switch controls the OS runtime libraries tracing. Multiple
APIs can be selected, separated by commas only (no spaces). Since OpenACC and cuXXX APIs
are tightly linked with CUDA, selecting one of those APIs will automatically enable CUDA tracing.
cublas, cudla, cusparse and cusolver all have XXX-verbose options available.
Reflex SDK latency markers will be automatically collected when DX or vulkan API trace is enabled.
See information on |
|
true, false |
false |
If true, trace any child process after fork and before they call one of the exec functions. Beware, tracing in this interval relies on undefined behavior and might cause your application to crash or deadlock. Note: This option is only available on Linux target platforms. |
|
|
true, false |
false |
Collect vsync events. If collection of vsync events is enabled, display/display_scanline ftrace events will also be captured. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
true, false, individual, batch, none |
individual |
If individual or true, trace each Vulkan workload’s GPU activity individually. If batch, trace
Vulkan workloads’ GPU activity in vkQueueSubmit call batches. If none or false, do not trace Vulkan
workloads’ GPU activity. Note that this switch is applicable only when |
|
|
primary,all |
all |
If primary, the CLI will wait on the application process termination. If all, the CLI will additionally wait on re-parented processes created by the application. |
|
|
true, false |
true |
If true, collect additional range of ETW events, including context status, allocations, sync wait
and signal events, etc. Note that this switch is applicable only when |
|
|
true, false |
false |
If true, collect backtraces of WDDM events. Disabling this data collection can reduce overhead for
certain target applications. Note that this switch is applicable only when |
|
|
< filepath pct.json > |
none |
Collect hypervisor trace. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
all, none, core, sched, irq, trap |
all |
Available in Nsight Systems Embedded Platforms Edition only. |
CLI Analyze Command Switch Options
The nsys analyze command generates and outputs to the terminal a report using expert system rules on existing results. Reports are generated from an SQLite export of a .nsys-rep file. If a .nsys-rep file is specified, Nsight Systems will look for an accompanying SQLite file and use it. If no SQLite export file exists, one will be created.
After choosing the analyze command switch, the following options are available. Usage:
nsys [global-options] analyze [options] [input-file]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
-f |
|
column, table, csv, tsv, json, hdoc, htable, . |
Specify the output format. The special name “.” indicates the default format for the given output. The default format for console is column, while files and process outputs default to csv. This option may be used multiple times. Multiple formats may also be specified using a comma-separated list (<name[:args…][,name[:args…]…]>). See Report Scripts for options available with each format. |
|
|
true, false |
false |
Force a re-export of the SQLite file from the specified .nsys-rep file, even if an SQLite file already exists. |
|
|
true, false |
false |
Overwrite any existing output files. |
|
|
<format_name>, ALL, [none] |
none |
With no argument, list a summary of the available output formats. If a format name is given, a more detailed explanation of the the format is displayed. If |
|
|
<rule_name>, ALL, [none] |
none |
With no argument, list available rules with a short description. If a rule name is given, a more detailed explanation of the rule is displayed. If |
|
-o |
|
-, @<command>, <basename>, . |
|
Specify the output mechanism. There are three output mechanisms: print to console, output to file, or output to command. This option may be used multiple times. Multiple outputs may also be specified using a comma-separated list. If the given output name is “-”, the output will be displayed on the console. If the output name starts with “@”, the output designates a command to run. The nsys command will be executed and the analysis output will be piped into the command. Any other output is assumed to be the base path and name for a file. If a file basename is given, the filename used will be: <basename>_<analysis&args>.<output_format>. The default base (including path) is the name of the SQLite file (as derived from the input file or |
-q |
|
Do not display verbose messages, only display errors. |
||
-r |
|
cuda_memcpy_async, cuda_memcpy_sync, cuda_memset_sync, cuda_api_sync, gpu_gaps, gpu_time_util, dx12_mem_ops |
all |
Specify the rules(s) to execute, including any arguments. This option may be used multiple times. Multiple rules may also be specified using a comma-separated list. See Expert Systems section and |
|
<file.sqlite> |
Specify the SQLite export filename. If this file exists, it will be used. If this file doesn’t exist (or if |
||
|
nsec, nanoseconds, usec, microseconds, msec, milliseconds, seconds |
nanoseconds |
Set basic unit of time. The argument of the switch is matched by using the longest prefix matching. Meaning that it is not necessary to write a whole word as the switch argument. It is similar to passing a “:time=<unit>” argument to every formatter, although the formatter uses more strict naming conventions. See |
CLI Cancel Command Switch Options
After choosing the cancel command switch, the following options are available. Usage:
nsys [global-options] cancel [options]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
|
<session identifier> |
none |
Cancel the collection in the given session. The option argument must represent a valid session name or ID as reported by |
CLI Export Command Switch Options
After choosing the export command switch, the following options are available. Usage:
nsys [global-options] export [options] [nsys-rep-file]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
-f |
|
true, false |
false |
If true, overwrite all existing result files with same output filename (nsys-rep, SQLITE, HDF, TEXT, ARROW, JSON). |
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
-l |
|
true, false |
true |
Controls if table creation is lazy or not. When true, a table will only be created when it contains data. This option will be deprecated in the future, and all exports will be non-lazy. This affects SQLite, HDF5, and Arrow exports only. |
-o |
|
<filename> |
<inputfile.ext> |
Set the .output filename. The default is the input filename with the extension for the chosen format. |
-q |
|
true, false |
false |
If true, do not display progress bar |
|
true,false |
false |
Output stored strings and thread names separately, with one value per line. This affects JSON and text output only. |
|
-t |
|
arrow, hdf, info json, sqlite, text |
sqlite |
Export format type. HDF format is supported only on x86_64 Linux and Windows |
|
true, false |
false |
If true, all timestamp values in the report will be shifted to UTC
wall-clock time, as defined by the UNIX epoch. This option can be used
in conjunction with the |
|
|
signed integer, in nanoseconds |
0 |
If given, all timestamp values in the report will be shifted by the given
amount. This option can be used in conjunction with the |
CLI Launch Command Switch Options
After choosing the launch command switch, the following options are available. Usage:
nsys [global-options] launch [options] <application> [application-arguments]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
-b |
|
auto,fp,lbr,dwarf,none |
Select the backtrace method to use while sampling. The option ‘lbr’ uses Intel(c) Corporation’s Last Branch Record registers, available only with Intel(c) CPUs codenamed Haswell and later. The option ‘fp’ is frame pointer and assumes that frame pointers were enabled during compilation. The option ‘dwarf’ uses DWARF’s CFI (Call Frame Information). Setting the value to ‘none’ can reduce collection overhead. |
|
|
true, false |
false |
Collect clock frequency changes. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
0x16, 0x17, …, none |
none |
Collect per-cluster Uncore PMU counters. Multiple values can be selected, separated by commas only (no
spaces). Use the |
|
|
< filename > |
none |
Open a file that contains profile switches and parse the switches. Note additional switches on the command line will override switches in the file. This flag can be specified more than once. |
|
|
0x11,0x13,…,none |
none |
Collect per-core PMU counters. Multiple values can be selected, separated by commas only (no spaces). Use
the |
|
|
0,1,2,…,none |
none |
Collect metrics on the CPU core. Multiple values can be selected, separated by commas only (no spaces). Use
the |
|
|
0x2a, 0x2c, …, none |
none |
Collect per-socket Uncore PMU counters. Multiple values can be selected, separated by commas only (no spaces).
Use the |
|
|
process-tree, system-wide, none |
process-tree |
Trace OS thread scheduling activity. Select ‘none’ to disable tracing CPU context switches. Depending on
the platform, some values may require admin or root privileges. Note: if the |
|
|
milliseconds |
See Description |
Set the interval, in milliseconds, when buffered CUDA data is automatically saved to storage. CUDA data buffer saves may cause profiler overhead. Buffer save behavior can be controlled with this switch. If the CUDA flush interval is set to 0 on systems running CUDA 11.0 or newer, buffers are saved when they fill. If a flush interval is set to a non-zero value on such systems, buffers are saved only when the flush interval expires. If a flush interval is set and the profiler runs out of available buffers before the flush interval expires, additional buffers will be allocated as needed. In this case, setting a flush interval can reduce buffer save overhead but increase memory use by the profiler. If the flush interval is set to 0 on systems running older versions of CUDA, buffers are saved at the end of the collection. If the profiler runs out of available buffers, additional buffers are allocated as needed. If a flush interval is set to a non-zero value on such systems, buffers are saved when the flush interval expires. A cuCtxSynchronize call may be inserted into the workflow before the buffers are saved which will cause application overhead. In this case, setting a flush interval can reduce memory use by the profiler but may increase save overhead. For collections over 30 seconds an interval of 10 seconds is recommended. Default is 10000 for Nsight Systems Embedded Platforms Edition and 0 otherwise. |
|
|
true, false |
false |
Track the GPU memory usage by CUDA kernels. Applicable only when CUDA tracing is enabled. Note: This feature may cause significant runtime overhead. |
|
|
true, false |
false |
This switch tracks the page faults that occur when CPU code tries to access a memory page that resides on the device. Note that this feature may cause significant runtime overhead. Not available on Nsight Systems Embedded Platforms Edition. |
|
|
true, false |
false |
This switch tracks the page faults that occur when GPU code tries to access a memory page that resides on the host. Note that this feature may cause significant runtime overhead. Not available on Nsight Systems Embedded Platforms Edition. |
|
|
all, none, kernel, memory, sync, other |
none |
When tracing CUDA APIs, enable the collection of a backtrace when a CUDA API is invoked. Significant runtime overhead may occur. Values may be combined using ‘,’. Each value except ‘none’ may be appended with a threshold after ‘:’. Threshold is duration, in nanoseconds, that CUDA APIs must execute before backtraces are collected, e.g. ‘kernel:500’. Default value for each threshold is 1000ns (1us). Note: CPU sampling must be enabled. |
|
|
graph, node |
graph |
If ‘graph’ is selected, CUDA graphs will be traced as a whole and node activities will not be collected. This will reduce overhead to a minimum, but requires CUDA driver version 515.43 or higher. If ‘node’ is selected, node activities will be collected, but CUDA graphs will not be traced as a whole. This may cause significant runtime overhead. Default is ‘graph’ if available, otherwise default is ‘node’. |
|
|
true, false |
false |
The Nsight Systems trace initialization involves creating a D3D device and discarding it. Enabling this flag makes a call to DXGIDeclareAdapterRemovalSupport() before device creation. Requires DX11 or DX12 trace to be enabled. |
|
|
true, false, individual, batch, none |
individual |
If individual or true, trace each DX12 workload’s GPU activity individually. If batch, trace DX12 workloads’
GPU activity in ExecuteCommandLists call batches. If none or false, do not trace DX12 workloads’ GPU
activity. Note that this switch is applicable only when |
|
|
true, false |
true |
If true, trace wait calls that block on fences for DX12. Note that this switch is applicable only when
|
|
-e |
|
A=B |
NA |
Set environment variable(s) for the application process to be launched. Environment variables should be defined as A=B. Multiple environment variables can be specified as A=B,C=D. |
|
help, <id1,id2,…>, all, none |
none |
Analyze video devices. |
|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
|
‘F1’ to ‘F12’ |
‘F12’ |
Hotkey to trigger the profiling session. Note that this switch is applicable only when
|
|
-n |
|
true, false |
true |
When true, the current environment variables and the tool’s environment variables will be specified for the launched process. When false, only the tool’s environment variables will be specified for the launched process. |
|
true,false |
true |
Use detours for injection. If false, process injection will be performed by windows hooks which allows it to bypass anti-cheat software. |
|
|
true, false |
false |
Trace Interrupt Service Routines (ISRs) and Deferred Procedure Calls (DPCs). Requires administrative privileges. Available only on Windows devices. |
|
|
openmpi,mpich |
openmpi |
When using |
|
-p |
|
range@domain, range, range@* |
none |
Specify NVTX range and domain to trigger the profiling session. This option is applicable only when
used along with |
|
default, <domain_names> |
Choose to exclude NVTX events from a comma separated list of domains. ‘default’ excludes NVTX events
without a domain. A domain with this name or commas in a domain name must be escaped with ‘\’.
Note: Only one of |
||
|
default, <domain_names> |
Choose to only include NVTX events from a comma separated list of domains. ‘default’ filters the
NVTX default domain. A domain with this name or commas in a domain name must be escaped with ‘\’.
Note: Only one of |
||
|
<json_file> |
Specify the path to the JSON file containing the requested functions to trace. |
||
|
true, false |
true |
If true, trace the OpenGL workloads’ GPU activity. Note that this switch is applicable only when
|
|
|
‘help’ or the end users selected events in the format ‘x,y’ |
Select the OS events to sample. Use the |
||
|
integer |
24 |
Set the depth for the backtraces collected for OS runtime libraries calls. |
|
|
integer |
6144 |
Set the stack dump size, in bytes, to generate backtraces for OS runtime libraries calls. |
|
|
nanoseconds |
80000 |
Set the duration, in nanoseconds, that all OS runtime libraries calls must execute before backtraces are collected. |
|
|
< nanoseconds > |
1000 ns |
Set the duration, in nanoseconds, that Operating System Runtime (osrt) APIs must execute before they are traced. Values significantly less than 1000 may cause significant overhead and result in extremely large result files. |
|
|
cuda, none, false |
none |
Collect Python backtrace event when tracing the selected API’s trigger. This option is supported
on Arm server (SBSA) platforms and x86 Linux targets. Note: the selected API tracing must be
enabled. For example, |
|
|
true, false |
false |
Collect Python backtrace sampling events. This option is supported on Arm server (SBSA) platforms, x86 Linux and Windows targets. Note: When profiling Python-only workflows, consider disabling the CPU sampling option to reduce overhead. |
|
|
1 < integers < 2000 |
1000 |
Specify the Python sampling frequency. The minimum supported frequency is 1Hz. The maximum
supported frequency is 2KHz. This option is ignored if the |
|
|
class/event,event, class/event:mode, class:mode,help,none |
none |
Multiple values can be selected, separated by commas only (no spaces). See the
|
|
|
system,process,fast,wide |
system:fast |
Values are separated by a colon (‘:’) only (no spaces). ‘system’ and ‘process’ cannot be specified at the same time. ‘fast’ and ‘wide’ cannot be specified at the same time. Please check the QNX documentation to determine when to select the ‘fast’ or ‘wide’ mode. Specify the default mode for QNX kernel events collection. |
|
|
true,false |
true |
Resolve symbols of captured samples and backtraces. |
|
|
true, false |
false |
Retain ETW files generated by the trace, merge and move the files to the output directory. |
|
|
< username > |
none |
Run the target application as the specified username. If not specified, the target application will be run by the same user as Nsight Systems. Requires root privileges. Available for Linux targets only. |
|
-s |
|
process-tree, system-wide, none |
process-tree |
Select how to collect CPU IP/backtrace samples. If ‘none’ is selected, CPU sampling is disabled.
Depending on the platform, some values may require admin or root privileges. If a target application
is launched, the default is ‘process-tree’, otherwise, the default is ‘none’. Note: ‘system-wide’
is not available on all platforms. Note: If set to ‘none’, CPU context switch data will still be
collected unless the |
|
integer <= 32 |
1 |
The number of CPU IP samples collected for every CPU IP/backtrace sample collected. For example, if set to 4, on the fourth CPU IP sample collected, a backtrace will also be collected. Lower values increase the amount of data collected. Higher values can reduce collection overhead and reduce the number of CPU IP samples dropped. If DWARF backtraces are collected, the default is 4, otherwise the default is 1. This option is not available on Nsight Systems Embedded Platforms Edition or on non-Linux targets. |
|
|
100 < integers < 8000 |
1000 |
Specify the sampling/backtracing frequency. The minimum supported frequency is 100 Hz. The maximum supported frequency is 8000 Hz. This option is supported only on QNX, Linux for Tegra, and Windows targets. |
|
|
integer |
determined dynamically |
The number of CPU Cycle events counted before a CPU instruction pointer (IP) sample is collected.
If configured, backtraces may also be collected. The smaller the sampling period, the higher the
sampling rate. Note that smaller sampling periods will increase overhead and significantly increase
the size of the result file(s). Requires |
|
|
integer |
determined dynamically |
The number of events counted before a CPU instruction pointer (IP) sample is collected. The event used to trigger the collection of a sample is determined dynamically. For example, on Intel based platforms, it will probably be “Reference Cycles” and on AMD platforms, “CPU Cycles”. If configured, backtraces may also be collected. The smaller the sampling period, the higher the sampling rate. Note that smaller sampling periods will increase overhead and significantly increase the size of the result file(s). This option is available only on Linux targets. |
|
|
timer, sched, perf, cuda |
timer,sched |
Specify backtrace collection trigger. Multiple APIs can be selected, separated by commas only (no spaces). Available on Nsight Systems Embedded Platforms Edition targets only. |
|
|
session identifier |
none |
Launch the application in the indicated session. The option argument must represent a valid session
name or ID as reported by |
|
|
[a-Z][0-9,a-Z,spaces] |
profile-<id>-<application> |
Name the session created by the command. Name must start with an alphabetical character followed by
printable or space characters. Any |
|
-w |
|
true, false |
true |
If true, send target process’s stdout and stderr streams to both the console and stdout/stderr files which are added to the report file. If false, only send target process stdout and stderr streams to the stdout/stderr files which are added to the report file. |
-t |
|
cuda, nvtx, cublas, cublas-verbose, cusparse, cusparse-verbose, cudnn, cudla, cudla-verbose, cusolver, cusolver-verbose, opengl, opengl-annotations, openacc, openmp, osrt, mpi, nvvideo, vulkan, vulkan-annotations, dx11, dx11-annotations, dx12, dx12-annotations, openxr, openxr-annotations, oshmem, ucx, wddm, tegra-accelerators, python-gil, syscall, none |
cuda, opengl, nvtx, osrt |
Select the API(s) to be traced. The osrt switch controls the OS runtime libraries tracing. Multiple
APIs can be selected, separated by commas only (no spaces). Since OpenACC and cuXXX APIs
are tightly linked with CUDA, selecting one of those APIs will automatically enable CUDA tracing.
cublas, cudla, cusparse and cusolver all have XXX-verbose options available.
Reflex SDK latency markers will be automatically collected when DX or vulkan API trace is enabled.
See information on |
|
true, false |
false |
If true, trace any child process after fork and before they call one of the exec functions. Beware, tracing in this interval relies on undefined behavior and might cause your application to crash or deadlock. Note: This option is only available on Linux target platforms. |
|
|
true, false, individual, batch, none |
individual |
If individual or true, trace each Vulkan workload’s GPU activity individually. If batch, trace
Vulkan workloads’ GPU activity in vkQueueSubmit call batches. If none or false, do not trace Vulkan
workloads’ GPU activity. Note that this switch is applicable only when |
|
|
primary,all |
all |
If primary, the CLI will wait on the application process termination. If all, the CLI will additionally wait on re-parented processes created by the application. |
|
|
true, false |
true |
If true, collect additional range of ETW events, including context status, allocations, sync wait
and signal events, etc. Note that this switch is applicable only when |
|
|
true, false |
false |
If true, collect backtraces of WDDM events. Disabling this data collection can reduce overhead for
certain target applications. Note that this switch is applicable only when |
CLI Sessions Command Switch Subcommands
After choosing the sessions command switch, the following subcommands are available. Usage:
nsys [global-options] sessions [subcommand]
Subcommand |
Description |
|---|---|
list |
List all active sessions including ID, name, and state information |
CLI Sessions List Command Switch Options
After choosing the sessions list command switch, the following options are available. Usage:
nsys [global-options] sessions list [options]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
-p |
|
true, false |
true |
Controls whether a header should appear in the output. |
CLI Shutdown Command Switch Options
After choosing the shutdown command switch, the following options are available. Usage:
nsys [global-options] shutdown [options]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
|
On Linux: one, sigkill, sigterm, signal number On Windows: true, false |
On Linux: sigterm On Windows: true |
Send signal to the target application’s process group when shutting down session. |
|
|
session identifier |
none |
Shutdown the indicated session. The option argument must represent a valid session name or ID as reported by |
CLI Start Command Switch Options
After choosing the start command switch, the following options are available. Usage:
nsys [global-options] start [options]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
none,tegra-accelerators |
none |
Collect other accelerators workload trace from the hardware engine units. Available in Nsight Systems Embedded Platforms Edition only. |
|
-b |
|
auto,fp,lbr,dwarf,none |
Select the backtrace method to use while sampling. The option ‘lbr’ uses Intel(c) Corporation’s Last Branch Record registers, available only with Intel(c) CPUs codenamed Haswell and later. The option ‘fp’ is frame pointer and assumes that frame pointers were enabled during compilation. The option ‘dwarf’ uses DWARF’s CFI (Call Frame Information). Setting the value to ‘none’ can reduce collection overhead. |
|
-c |
|
none, cudaProfilerApi, hotkey, nvtx |
none |
When |
|
none, stop, stop-shutdown, repeat[:N], repeat-shutdown:N |
stop-shutdown |
Specify the desired behavior when a capture range ends. Applicable only when used along with
|
|
|
‘help’ or the end users selected events in the format ‘x,y’ |
‘2’ i.e. Instructions Retired |
Select the CPU Core events to sample. Use the |
|
|
process-tree, system-wide, none |
process-tree |
Trace OS thread scheduling activity. Select ‘none’ to disable tracing CPU context switches. Depending on
the platform, some values may require admin or root privileges. Note: if the |
|
|
<plugin_name>[,arg1,arg2,…] |
NA |
Use the specified plugin. The option can be specified multiple times to enable multiple plugins.
Plugin arguments are separated by commas only (no spaces). Commas can be escaped with a backslash ‘\’.
The backslash itself can be escaped by another backslash ‘\\’. To include spaces in an argument,
enclose the argument in double quotes ‘”’. To list all available plugins, use |
|
|
true, false |
false |
Enable EL1 sampling. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
< filepath config.json > |
none |
EL1 sampling config. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
“<name>,<guid>”, or path to JSON file |
none |
Add custom ETW trace provider(s). If you want to specify more attributes than Name and GUID, provide a JSON configuration file as as outlined below. This switch can be used multiple times to add multiple providers. Note: Only available for Windows targets. |
|
|
system-wide, none |
none |
Use the |
|
|
Integers from 1 to 20 Hz |
3 |
The sampling frequency used to collect event counts. Minimum event sampling frequency is 1 Hz. Maximum event sampling frequency is 20 Hz. Not available in Nsight Systems Embedded Platforms Edition. |
|
|
arrow, hdf, json, sqlite, text, none |
none |
Create additional output file(s) based on the data collected. This option can be given more than once. WARNING: If the collection captures a large amount of data, creating the export file may take several minutes to complete. |
|
|
true, false |
true |
If set to true, any call to cudaProfilerStop() will cause the CUDA trace buffers to be flushed. Note that the CUDA trace buffers will be flushed when the collection ends, irrespective of the value of this switch. |
|
-f |
|
true, false |
false |
If true, overwrite all existing result files with same output filename (.nsys-rep, .arrows, .h5, .json, .sqlite, .txt). |
|
Collect ftrace events. Argument should list events to collect as: subsystem1/event1,subsystem2/event2. Requires root. No ftrace events are collected by default. |
|||
|
Skip initial ftrace setup and collect already configured events. Default resets the ftrace configuration. |
|||
|
GPU ID, help, all, none |
none |
Collect GPU Metrics from specified devices. Determine GPU IDs by using |
|
|
integer |
10000 |
Specify GPU Metrics sampling frequency. Minimum supported frequency is 10 (Hz). Maximum supported frequency is 200000 (Hz). |
|
|
alias, file:<file name> |
see description |
Specify metric set for GPU Metrics. The argument must be one of the aliases reported by
|
|
|
help, <id1,id2,…>, all, none |
none |
Analyze video devices. |
|
|
true,false |
false |
Trace GPU context switches. Note that this requires driver r435.17 or later and root permission. |
|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
|
<NIC names> |
none |
A comma-separated list of NIC names. The NICs which |
|
|
<file paths> |
none |
A comma-separated list of file paths. Paths of an existing ibdiagnet db_csv files, containing networks
information data. Nsight Systems will read the networks information from these files. Don’t use |
|
|
<directory path> |
none |
Sets the path of a directory into which ibdiagnet network discovery data will be written. Use this option
together with the |
|
|
<IB switch GUIDs> |
none |
A comma-separated list of InfiniBand switch GUIDs. Collect InfiniBand switch congestion events from switches identified by the specified GUIDs. This switch can be used multiple times. System scope. Use the –ib-switch-congestion-nic-device, –ib-switch-congestion-percent, and –ib-switch-congestion-threshold-high switches to further control how congestion events are collected. |
|
|
<NIC name> |
none |
The name of the NIC (HCA) through which InfiniBand switches will be accessed. By default, the first active NIC will be used. One way to find a NIC’s name is via the ibnetdiscover –Hca_list | grep “$(hostname)” command. Example usage: –ib-switch-congestion-nic-device=mlx5_3 |
|
|
1 <= integer <= 100 |
50 |
Percent of InfiniBand switch congestion events to be collected. This option enables reducing the network bandwidth consumed by reporting congestion events. |
|
|
1 <= integer <= 1023 |
75 |
High threshold percentage for InfiniBand switch egress port buffer size. Before a packet leaves an InfiniBand switch, it is stored at an egress port buffer. The buffer’s size is checked and if it exceeds the given threshold percentage, a congestion event is reported. The percentage can be greater than 100. |
|
|
true, false |
false |
Trace Interrupt Service Routines (ISRs) and Deferred Procedure Calls (DPCs). Requires administrative privileges. Available only on Windows devices. |
|
|
true, false |
false |
Collect metrics from supported NIC/HCA devices. System scope. Not available on Nsight Systems Embedded Platforms Edition. |
|
|
‘help’ or the end users selected events in the format ‘x,y’ |
Select the OS events to sample. Use the |
||
-o |
|
< filename > |
report# |
Set report file name. Any %q{ENV_VAR} pattern in the filename will be substituted with the value of the environment variable. Any %h pattern in the filename will be substituted with the hostname of the system. Any %p pattern in the filename will be substituted with the PID of the target process or the PID of the root process if there is a process tree. Any %% pattern in the filename will be substituted with %. Default is report#.{nsys-rep,sqlite,h5,txt,arrows,json} in the working directory. |
|
main, process-tree, system-wide |
main |
Select which process(es) to trace. Available in Nsight Systems Embedded Platforms Edition only. Nsight Systems Workstation Edition will always trace system-wide in this version of the tool. |
|
|
true, false |
false |
Retain ETW files generated by the trace, merge and move the files to the output directory. |
|
-s |
|
process-tree, system-wide, none |
process-tree |
Select how to collect CPU IP/backtrace samples. If ‘none’ is selected, CPU sampling is disabled.
Depending on the platform, some values may require admin or root privileges. If a target application
is launched, the default is ‘process-tree’, otherwise, the default is ‘none’. Note: ‘system-wide’
is not available on all platforms. Note: If set to ‘none’, CPU context switch data will still be
collected unless the |
|
integer <= 32 |
1 |
The number of CPU IP samples collected for every CPU IP/backtrace sample collected. For example, if set to 4, on the fourth CPU IP sample collected, a backtrace will also be collected. Lower values increase the amount of data collected. Higher values can reduce collection overhead and reduce the number of CPU IP samples dropped. If DWARF backtraces are collected, the default is 4, otherwise the default is 1. This option is not available on Nsight Systems Embedded Platforms Edition or on non-Linux targets. |
|
|
100 < integers < 8000 |
1000 |
Specify the sampling/backtracing frequency. The minimum supported frequency is 100 Hz. The maximum supported frequency is 8000 Hz. This option is supported only on QNX, Linux for Tegra, and Windows targets. |
|
|
integer |
determined dynamically |
The number of CPU Cycle events counted before a CPU instruction pointer (IP) sample is collected.
If configured, backtraces may also be collected. The smaller the sampling period, the higher the
sampling rate. Note that smaller sampling periods will increase overhead and significantly increase
the size of the result file(s). Requires |
|
|
integer |
determined dynamically |
The number of events counted before a CPU instruction pointer (IP) sample is collected. The event used to trigger the collection of a sample is determined dynamically. For example, on Intel based platforms, it will probably be “Reference Cycles” and on AMD platforms, “CPU Cycles”. If configured, backtraces may also be collected. The smaller the sampling period, the higher the sampling rate. Note that smaller sampling periods will increase overhead and significantly increase the size of the result file(s). This option is available only on Linux targets. |
|
|
timer, sched, perf, cuda |
timer,sched |
Specify backtrace collection trigger. Multiple APIs can be selected, separated by commas only (no spaces). Available on Nsight Systems Embedded Platforms Edition targets only. |
|
|
[a-Z][0-9,a-Z,spaces] |
profile-<id>-<application> |
Name the session created by the command. Name must start with an alphabetical character followed by
printable or space characters. Any |
|
-w |
|
true, false |
true |
If true, send target process’s stdout and stderr streams to both the console and stdout/stderr files which are added to the report file. If false, only send target process stdout and stderr streams to the stdout/stderr files which are added to the report file. |
|
true,false |
false |
Collect SoC Metrics. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
integer |
10000 |
Specify SoC Metrics sampling frequency. Minimum supported frequency is ‘100’ (Hz). Maximum supported frequency is ‘1000000’ (Hz). Available in Nsight Systems Embedded Platforms Edition only. |
|
|
alias, file:<file name> |
see description |
Specify metric set for SoC Metrics. The argument must be one of the aliases reported by
|
|
|
true, false |
false |
Generate summary statistics after the collection. WARNING: When set to true, an SQLite database will be created after the collection. If the collection captures a large amount of data, creating the database file may take several minutes to complete. |
|
-x |
|
true, false |
true |
If true, stop collecting automatically when the launched process has exited or when the duration expires - whichever occurs first. If false, duration must be set and the collection stops only when the duration expires. Nsight Systems does not officially support runs longer than 5 minutes. |
|
true, false |
false |
Collect vsync events. If collection of vsync events is enabled, display/display_scanline ftrace events will also be captured. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
< filepath pct.json > |
none |
Collect hypervisor trace. Available in Nsight Systems Embedded Platforms Edition only. |
|
|
all, none, core, sched, irq, trap |
all |
Available in Nsight Systems Embedded Platforms Edition only. |
CLI Stats Command Switch Options
The nsys stats command generates a series of summary or trace reports.
These reports can be output to the console, or to individual files, or piped to
external processes. Reports can be rendered in a variety of different output
formats, from human readable columns of text, to formats more appropriate for
data exchange, such as CSV.
Reports are generated from an SQLite export of a .nsys-rep file. If a .nsys-rep file is specified, Nsight Systems will look for an accompanying SQLite file and use it. If no SQLite file exists, one will be exported and created.
Individual reports are generated by calling out to scripts that read data from the SQLite file and return their report data in CSV format. Nsight Systems ingests this data and formats it as requested, then displays the data to the console, writes it to a file, or pipes it to an external process. Adding new reports is as simple as writing a script that can read the SQLite file and generate the required CSV output. See the shipped scripts as an example. Both reports and formatters may take arguments to tweak their processing. For details on shipped scripts and formatters, see Report Scripts topic.
Reports are processed using a three-tuple that consists of 1) the requested report (and any arguments), 2) the presentation format (and any arguments), and 3) the output (filename, console, or external process). The first report specified uses the first format specified, and is presented via the first output specified. The second report uses the second format for the second output, and so forth. If more reports are specified than formats or outputs, the format and/or output list is expanded to match the number of provided reports by repeating the last specified element of the list (or the default, if nothing was specified).
nsys stats is a very powerful command and can handle complex argument
structures, please see the topic below on Example Stats Command Sequences.
After choosing the stats command switch, the following options are
available. Usage:
nsys [global-options] stats [options] [input-file]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
-f |
|
column, table, csv, tsv, json, hdoc, htable, . |
Specify the output format. The special name “.” indicates the default format for the given output. The default format for console is column, while files and process outputs default to csv. This option may be used multiple times. Multiple formats may also be specified using a comma-separated list (<name[:args…][,name[:args…]…]>). See Report Scripts for options available with each format. |
|
|
true, false |
false |
Force a re-export of the SQLite file from the specified .nsys-rep file, even if an SQLite file already exists. |
|
|
true, false |
false |
Overwrite any existing report file(s). |
|
|
<format_name>, ALL, [none] |
none |
With no argument, give a summary of the available output formats. If a format name is given, a more detailed explanation of that
format is displayed. If |
|
|
<report_name>, ALL, [none] |
none |
With no argument, list a summary of the available summary and trace reports. If a report name is given, a more detailed explanation
of the report is displayed. If |
|
-o |
|
-, @<command>, <basename>, . |
|
Specify the output mechanism. There are three output mechanisms: print to console, output to file, or output to command. This
option may be used multiple times. Multiple outputs may also be specified using a comma-separated list. If the given output name
is “-”, the output will be displayed on the console. If the output name starts with “@”, the output designates a command to run.
The nsys command will be executed and the analysis output will be piped into the command. Any other output is assumed to be the
base path and name for a file. If a file basename is given, the filename used will be: <basename>_<analysis&args>.<output_format>.
The default base (including path) is the name of the SQLite file (as derived from the input file or |
-q |
|
Do not display verbose messages, only display errors. |
||
-r |
|
See Report Scripts |
Specify the report(s) to generate, including any arguments. This option may be used multiple times. Multiple reports may also be specified using a comma-separated list (<name[:args…][,name[:args…]…]>). If no reports are specified, the following will be used as the default report set: nvtx_sum, osrt_sum, cuda_api_sum, cuda_gpu_kern_sum, cuda_gpu_mem_time_sum, cuda_gpu_mem_size_sum, openmp_sum, opengl_khr_range_sum, opengl_khr_gpu_range_sum, vulkan_marker_sum, vulkan_gpu_marker_sum, dx11_pix_sum, dx12_gpu_marker_sum, dx12_pix_sum, wddm_queue_sum, um_sum, um_total_sum, um_cpu_page_faults_sum, openacc_sum. See Report Scripts section for details about existing built-in scripts and how to make your own. |
|
|
<path> |
Add a directory to the path used to find report scripts. This is usually only needed if you have one or more directories with
personal scripts. This option may be used multiple times. Each use adds a new directory to the end of the path. A search path can
also be defined using the environment variable “NSYS_STATS_REPORT_PATH”. Directories added this way will be added after the
application flags. The last two entries in the path will always be the current working directory, followed by the directory
containing the shipped |
||
|
<file.sqlite> |
Specify the SQLite export filename. If this file exists, it will be used. If this file doesn’t exist (or if |
||
|
nsec, nanoseconds, usec, microseconds, msec, milliseconds, seconds |
nanoseconds |
Set basic unit of time. The argument of the switch is matched by using the longest prefix matching. Meaning that it is not
necessary to write a whole word as the switch argument. It is similar to passing a “:time=<unit>” argument to every formatter,
although the formatter uses more strict naming conventions. See |
CLI Status Command Switch Options
The nsys status command returns the current state of the CLI. After choosing the status command switch, the following options are available. Usage:
nsys [global-options] status [options]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
Prints information for all the available profiling environments. |
|||
-e |
|
Returns information about the system regarding suitability of the profiling environment. |
||
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
-n |
|
Returns information about the system regarding suitability of the network profiling environment. |
||
|
session identifier |
none |
Print the status of the indicated session. The option argument must represent a valid session name or ID as reported by |
CLI Stop Command Switch Options
After choosing the stop command switch, the following options are available. Usage:
nsys [global-options] stop [options]
Short |
Long |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
<tag> |
none |
Print the help message. The option can take one optional argument that will be used as a tag. If a tag is provided, only options relevant to the tag will be printed. |
|
|
session identifier |
none |
Stop the indicated session. The option argument must represent a valid session name or ID as reported by |
Example Single Command Lines
Version Information
nsys -v
Effect: Prints tool version information to the screen.
Run with elevated privilege
sudo nsys profile <app>
Effect: Nsight Systems CLI (and target application) will run with elevated privilege. This is necessary for some features, such as FTrace or system-wide CPU sampling. If you don’t want the target application to be elevated, use --run-as option.
Default analysis run
nsys profile <application>
[application-arguments]
Effect: Launch the application using the given arguments. Start collecting immediately and end collection when the application stops. Trace CUDA, OpenGL, NVTX, and OS runtime libraries APIs. Collect CPU sampling information and thread scheduling information. With Nsight Systems Embedded Platforms Edition this will only analysis the single process. With Nsight Systems Workstation Edition this will trace the process tree. Generate the report#.nsys-rep file in the default location, incrementing the report number if needed to avoid overwriting any existing output files.
Limited trace only run
nsys profile --trace=cuda,nvtx -d 20
--sample=none --cpuctxsw=none -o my_test <application>
[application-arguments]
Effect: Launch the application using the given arguments. Start collecting immediately and end collection after 20 seconds or when the application ends. Trace CUDA and NVTX APIs. Do not collect CPU sampling information or thread scheduling information. Profile any child processes. Generate the output file as my_test.nsys-rep in the current working directory.
Delayed start run
nsys profile -e TEST_ONLY=0 -y 20
<application> [application-arguments]
Effect: Set environment variable TEST_ONLY=0. Launch the application using the given arguments. Start collecting after 20 seconds and end collection at application exit. Trace CUDA, OpenGL, NVTX, and OS runtime libraries APIs. Collect CPU sampling and thread schedule information. Profile any child processes. Generate the report#.nsys-rep file in the default location, incrementing if needed to avoid overwriting any existing output files.
Collect ftrace events
nsys profile --ftrace=drm/drm_vblank_event
-d 20
Effect: Collect ftrace drm_vblank_event events for 20 seconds. Generate the report#.nsys-rep file in the current working directory. Note that ftrace event collection requires running as root. To get a list of ftrace events available from the kernel, run the following:
sudo cat /sys/kernel/debug/tracing/available_events
Run GPU metric sampling on one TU10x
nsys profile --gpu-metrics-devices=0
--gpu-metrics-set=tu10x-gfxt <application>
Effect: Launch application. Collect default options and GPU metrics for the first GPU (a TU10x), using the tu10x-gfxt metric set at the default frequency (10 kHz). Profile any child processes. Generate the report#.nsys-rep file in the default location, incrementing if needed to avoid overwriting any existing output files.
Run GPU metric sampling on all GPUs at a set frequency
nsys profile --gpu-metrics-devices=all
--gpu-metrics-frequency=20000 <application>
Effect: Launch application. Collect default options and GPU metrics for all available GPUs using the first suitable metric set for each and sampling at 20 kHz. Profile any child processes. Generate the report#.nsys-rep file in the default location, incrementing if needed to avoid overwriting any existing output files.
Collect CPU IP/backtrace and CPU context switch
nsys profile --sample=system-wide --duration=5
Effect: Collects both CPU IP/backtrace samples using the default backtrace mechanism and traces CPU context switch activity for the whole system for 5 seconds. Note that it requires root permission to run. No hardware or OS events are sampled. Post processing of this collection will take longer due to the large number of symbols to be resolved caused by system-wide sampling.
Get list of available CPU core events
nsys profile --cpu-core-events=help
Effect: Lists the CPU events that can be sampled and the maximum number of CPU events that can be sampled concurrently.
Collect system-wide CPU events and trace application
nsys profile --event-sample=system-wide
--cpu-core-events='1,2' --event-sampling-frequency=5 <app> [app args]
Effect:Collects CPU IP/backtrace samples using the default backtrace mechanism, traces CPU context switch activity, and samples each CPU’s “CPU Cycles” and “Instructions Retired” event every 200 ms for the whole system. Note that it requires root permission to run. Note that CUDA, NVTX, OpenGL, and OSRT within the app launched by Nsight Systems are traced by default while using this command. Post processing of this collection will take longer due to the large number of symbols to be resolved caused by system-wide sampling.
Collect custom ETW trace using configuration file
nsys profile --etw-provider=file.JSON
Effect: Configure custom ETW collectors using the contents of file.JSON. Collect data for 20 seconds. Generate the report#.nsys-rep file in the current working directory.
A template JSON configuration file is located at in the Nsight Systems installation directory as \target-windows-x64\etw_providers_template.json. This path will show up automatically if you call
nsys profile --help
The level attribute can only be set to one of the following:
TRACE_LEVEL_CRITICAL
TRACE_LEVEL_ERROR
TRACE_LEVEL_WARNING
TRACE_LEVEL_INFORMATION
TRACE_LEVEL_VERBOSE
The flags attribute can only be set to one or more of the following:
EVENT_TRACE_FLAG_ALPC
EVENT_TRACE_FLAG_CSWITCH
EVENT_TRACE_FLAG_DBGPRINT
EVENT_TRACE_FLAG_DISK_FILE_IO
EVENT_TRACE_FLAG_DISK_IO
EVENT_TRACE_FLAG_DISK_IO_INIT
EVENT_TRACE_FLAG_DISPATCHER
EVENT_TRACE_FLAG_DPC
EVENT_TRACE_FLAG_DRIVER
EVENT_TRACE_FLAG_FILE_IO
EVENT_TRACE_FLAG_FILE_IO_INIT
EVENT_TRACE_FLAG_IMAGE_LOAD
EVENT_TRACE_FLAG_INTERRUPT
EVENT_TRACE_FLAG_JOB
EVENT_TRACE_FLAG_MEMORY_HARD_FAULTS
EVENT_TRACE_FLAG_MEMORY_PAGE_FAULTS
EVENT_TRACE_FLAG_NETWORK_TCPIP
EVENT_TRACE_FLAG_NO_SYSCONFIG
EVENT_TRACE_FLAG_PROCESS
EVENT_TRACE_FLAG_PROCESS_COUNTERS
EVENT_TRACE_FLAG_PROFILE
EVENT_TRACE_FLAG_REGISTRY
EVENT_TRACE_FLAG_SPLIT_IO
EVENT_TRACE_FLAG_SYSTEMCALL
EVENT_TRACE_FLAG_THREAD
EVENT_TRACE_FLAG_VAMAP
EVENT_TRACE_FLAG_VIRTUAL_ALLOC
Typical case: profile a Python script that uses CUDA
nsys profile --trace=cuda,cudnn,cublas,osrt,nvtx
--delay=60 python my_dnn_script.py
Effect: Launch a Python script and start profiling it 60 seconds after the launch, tracing CUDA, cuDNN, cuBLAS, OS runtime APIs, and NVTX as well as collecting thread schedule information.
Typical case: profile an app that uses Vulkan
nsys profile --trace=vulkan,osrt,nvtx
--delay=60 ./myapp
Effect: Launch an app and start profiling it 60 seconds after the launch, tracing Vulkan, OS runtime APIs, and NVTX as well as collecting CPU sampling and thread schedule information.
Example Interactive CLI Command Sequences
Collect from beginning of application, end manually
nsys start --stop-on-exit=false
nsys launch --trace=cuda,nvtx --sample=none <application> [application-arguments]
nsys stop
Effect: Create interactive CLI process and set it up to begin collecting as soon as an application is launched. Launch the application, set up to allow tracing of CUDA and NVTX as well as collection of thread schedule information. Stop only when explicitly requested. Generate the report#.nsys-rep in the default location.
Note
If you start a collection and fail to stop the collection (or if you are allowing it to stop on exit, and the application runs for too long) your system’s storage space may be filled with collected data causing significant issues for the system. Nsight Systems will collect a different amount of data/sec depending on options, but in general Nsight Systems does not support runs of more than 5 minutes duration.
Run application, begin collection manually, run until process ends
nsys launch -w true <application> [application-arguments]
nsys start
Effect: Create interactive CLI and launch an application set up for default analysis. Send application output to the terminal. No data is collected until you manually start collection at area of interest. Profile until the application ends. Generate the report#.nsys-rep in the default location.
Note
If you launch an application and that application and any descendants exit before start is called Nsight Systems will create a fully formed .nsys-rep file containing no data.
Run application, start/stop collection using cudaProfilerStart/Stop
nsys start -c cudaProfilerApi
nsys launch -w true <application> [application-arguments]
Effect: Create interactive CLI process and set it up to begin collecting as soon
as a cudaProfileStart() is detected. Launch application for default analysis,
sending application output to the terminal. Stop collection at next call to
cudaProfilerStop, when the user calls nsys stop, or when the root process
terminates. Generate the report#.nsys-rep in the default location.
Note
If you call nsys launch before nsys start -c cudaProfilerApi and the
code contains a large number of short duration cudaProfilerStart/Stop pairs,
Nsight Systems may be unable to process them correctly, causing a fault. This
will be corrected in a future version.
Note
Use the Nsight Systems CLI option --capture-range-end-repeat to capture
a separate report file for each capture range defined by calls to
cudaProfilerStart/Stop. To avoid overwriting report files unexpectedly,
Nsight Systems will ignore the --force-overwrite option in this case.
Run application, start/stop collection using NVTX
nsys start -c nvtx
nsys launch -w true -p MESSAGE@DOMAIN <application> [application-arguments]
Effect: Create interactive CLI process and set it up to begin collecting as soon as an NVTX range with given message in given domain (capture range) is opened. Launch application for default analysis, sending application output to the terminal. Stop collection when all capture ranges are closed, when the user calls nsys stop, or when the root process terminates. Generate the report#.nsys-rep in the default location.
Note
The Nsight Systems CLI only triggers the profiling session for the first capture range.
NVTX capture range can be specified:
Message@Domain: All ranges with given message in given domain are capture ranges. For example:
nsys launch -w true -p profiler@service ./app
This would make the profiling start when the first range with message “profiler” is opened in domain “service”.
Message@*: All ranges with given message in all domains are capture ranges. For example:
nsys launch -w true -p profiler@* ./app
This would make the profiling start when the first range with message “profiler” is opened in any domain.
Message: All ranges with given message in default domain are capture ranges. For example:
nsys launch -w true -p profiler ./app
This would make the profiling start when the first range with message “profiler” is opened in the default domain.
By default only messages, provided by NVTX registered strings are considered to avoid additional overhead. To enable non-registered strings check please launch your application with
NSYS_NVTX_PROFILER_REGISTER_ONLY=0environment:nsys launch -w true -p profiler@service -e NSYS_NVTX_PROFILER_REGISTER_ONLY=0 ./app
Note
The separator ‘@’ can be escaped with backslash ‘\’. If multiple separators without escape character are specified, only the last one is applied, all others are discarded.
Run application, start/stop collection multiple times
The interactive CLI supports multiple sequential collections per launch.
nsys launch <application> [application-arguments]
nsys start
nsys stop
nsys start
nsys stop
nsys shutdown --kill sigkill
Effect: Create interactive CLI and launch an application set up for default analysis. Send application output to the terminal. No data is collected until the start command is executed. Collect data from start until stop requested, generate report#.qstrm in the current working directory. Collect data from second start until the second stop request, generate report#.nsys-rep (incremented by one) in the current working directory. Shutdown the interactive CLI and send sigkill to the target application’s process group.
Note
Calling nsys cancel after nsys start will cancel the collection without generating a report.
Example Stats Command Sequences
Display default statistics
nsys stats report1.nsys-rep
Effect: Export an SQLite file named report1.sqlite from report1.nsys-rep (assuming it does not already exist). Print the default reports in column format to the console.
Note: The following two command sequences should present very similar information:
nsys profile --stats=true <application>
or
nsys profile <application>
nsys stats report1.nsys-rep
Display specific data from a report
nsys stats --report cuda_gpu_trace report1.nsys-rep
Effect: Export an SQLite file named report1.sqlite from report1.nsys-rep (assuming it does not already exist). Print the report generated by the cuda_gpu_trace script to the console in column format.
Generate multiple reports, in multiple formats, output multiple places
nsys stats --report cuda_gpu_trace --report cuda_gpu_kern_sum --report cuda_api_sum --format csv,column --output .,- report1.nsys-rep
Effect: Export an SQLite file named report1.sqlite from report1.nsys-rep (assuming it does not already exist). Generate three reports. The first, the cuda_gpu_trace report, will be output to the file report1_cuda_gpu_trace.csv in CSV format. The other two reports, cuda_gpu_kern_sum and cuda_api_sum, will be output to the console as columns of data. Although three reports were given, only two formats and outputs are given. To reconcile this, both the list of formats and outputs is expanded to match the list of reports by repeating the last element.
Submit report data to a command
nsys stats --report cuda_api_sum --format table \ --output @“grep -E (-|Name|cudaFree” test.sqlite
Effect: Open test.sqlite and run the cuda_api_sum script on that file. Generate table data and feed that into the command grep -E (-|Name|cudaFree). The grep command will filter out everything but the header, formatting, and the cudaFree data, and display the results to the console.
Note: When the output name starts with @, it is defined as a command. The command is run, and the output of the report is piped to the command’s stdin (standard-input). The command’s stdout and stderr remain attached to the console, so any output will be displayed directly to the console.
Be aware there are some limitations in how the command string is parsed. No shell expansions (including *, ?, [], and ~) are supported. The command cannot be piped to another command, nor redirected to a file using shell syntax. The command and command arguments are split on whitespace, and no quotes (within the command syntax) are supported. For commands that require complex command line syntax, it is suggested that the command be put into a shell script file, and the script designated as the output command
Example Output from --stats Option
The nsys stats command can be used post analysis to generate specific or personalized reports. For a default fixed set of summary statistics to be automatically generated, you can use the --stats option with the nsys profile or nsys start command to generate a fixed set of useful summary statistics.
If your run traces CUDA, these include CUDA API, Kernel, and Memory Operation statistics:
If your run traces OS runtime events or NVTX push-pop ranges:
If your run traces graphics debug markers these include DX11 debug markers, DX12 debug markers, Vulkan debug markers or KHR debug markers:
Recipes for these statistics as well as documentation on how to create your own metrics will be available in a future version of the tool.
System Wide API Trace on Windows
On Windows, Nsight Systems can trace certain APIs (currently supported: DX11, DX12 and Vulkan) in already-running applications, by way of system-wide API trace from the CLI.
To initiate system-wide API tracing, run the Nsight Systems CLI with the
--trace option including one or more of the supported APIs, the
--system-wide``option set to ``true, and without specifying a target
application. System-wide API tracing may be combined with trace types that are
always system-wide such as --trace=wddm.
To trace a DX11 or DX12 target application, it must gain the system focus, the user must either click on the application window or using Alt+Tab to select it.
For example, to trace multiple DX12 applications with PIX markers and GPU workload trace as well as WDDM events for the next 20 seconds, run the command:
nsys profile --trace=dx12-annotations,wddm --dx12-gpu-workload=individual
--duration=20
then click each of the target applications’ windows to give them focus.
To trace a Vulkan target application, it must be launched after the nsys profile
command has been executed.
Opening Command Line Results Files for Visualization
Open CLI results in GUI
The CLI will generate an .nsys-rep file after analysis is complete. This file can be opened in any GUI that is the same version or a more recent version.
The opening of really large, multi-gigabyte, .nsys-rep files may take up all of the memory on the host computer and lock up the system. This will be fixed in a later version.
Importing Windows ETL files
For Windows targets, ETL files captured with Xperf or the log.cmd command
supplied with GPUView in the Windows Performance Toolkit can be imported to
create reports as if they were captured with Nsight Systems’s “WDDM trace” and
“Custom ETW trace” features. Simply choose the .etl file from the Import dialog
to convert it to a .nsys-rep file.
Handling Application Launchers (mpirun, deepspeed, etc)
Nsight Systems has built-in API trace support for various communication APIs,
such as MPI, OpenSHMEM, UCX, NCCL and NVSHMEM.
To execute respective programs on multiple different machines (compute nodes),
usually launchers are used, e.g. mpirun/mpiexec (MPI),
shmemrun/oshrun (OpenSHMEM),``srun`` (SLURM) or deepspeed.
In single-node profiling sessions, the Nsight Systems CLI can be prefixed before the program (binary) or the launcher. In the latter case, the execution of the launcher will also be profiled and all processes will be recorded into the same report file, e.g for mpirun
nsys profile [nsys args] mpirun [mpirun args] ...
In multi-node profiling sessions, the Nsight Systems CLI has to be prefixed before the application, but not before the launcher command, e.g. for mpirun
mpirun [mpirun args] nsys profile [nsys args] ...
You can use %q{OMPI_COMM_WORLD_RANK} (Open MPI), %q{PMI_RANK} (MPICH)
%q{SLURM_PROCID} (Slurm) or %p in the -o|--output option to include
the rank or process ID into the report file name.
Warning: An error will occur, if several processes want to write to the same report file at the same time.
Profile a Single Process or a Subset of Processes
It might be reasonable to profile only a single or a few representative processes of a program run, e.g. to reduce the amount of measurement data.
To achieve this, a wrapper script can be used. The following script called nsys_profile.sh uses nsys to profile MPI rank 0 only.
#!/bin/bash
# Use $PMI_RANK for MPICH and $SLURM_PROCID with srun.
if [ $OMPI_COMM_WORLD_RANK -eq 0 ]; then
nsys profile -e NSYS_MPI_STORE_TEAMS_PER_RANK=1 -t mpi "$@"
else
"$@"
fi
You can change the profiling options accordingly and execute with mpirun [mpirun options] ./nsys_profile.sh ./myapp [app options]. The above code can be easily adapted for OpenSHMEM and SLURM launchers.
Note
If only a subset of MPI ranks is profiled, set the environment variable
NSYS_MPI_STORE_TEAMS_PER_RANK=1 to store all members of custom MPI
communicators per MPI rank. Otherwise, the execution might hang or fail with
an MPI error.
DeepSpeed
DeepSpeed provides a parallel launcher, which launches a Python script on multiple nodes and/or GPUs. For multi-node runs, a simple wrapper script (nsys_profile.sh) is required to profile with Nsight Systems.
#! /bin/bash nsys profile -t cuda,mpi,nvtx,cudnn -o rname.%p python …
This above script has to be used with the –no-python option:
deepspeed –no_python [deepspeed args] ./nsys_profile.sh
GPU and NIC metrics collection:
If multiple instances of nsys profile are executed concurrently on the same
node and GPU and/or NIC metrics collection is enabled, each process will collect
metrics for all available NICs and tries to collect GPU metrics for the
specified devices. This can be avoided with a simple bash script similar to the
following:
#!/bin/bash
# Use $SLURM_LOCALID with srun.
if [ $OMPI_COMM_WORLD_LOCAL_RANK -eq 0 ]; then
nsys profile --nic-metrics=true --gpu-metrics-devices=all "$@"
else
nsys profile "$@"
fi
This above script will collect NIC and GPU metrics only for one rank, the node-local rank 0. Alternatively, if one rank per GPU is used, the GPU metrics devices can be specified based on the node-local rank in a wrapper script as follows:
#!/bin/bash
# Use $SLURM_LOCALID with srun.
nsys profile -e CUDA_VISIBLE_DEVICES=${OMPI_COMM_WORLD_LOCAL_RANK} \
--gpu-metrics-devices=${OMPI_COMM_WORLD_LOCAL_RANK} "$@"
Profiling from the GUI
Profiling Linux Targets from the GUI
Connecting to the Target Device
Nsight Systems provides a simple interface to profile on localhost or manage multiple connections to Linux or Windows based devices via SSH. The network connections manager can be launched through the device selection dropdown:
On x86_64:
On Tegra:
The dialog has simple controls that allow adding, removing, and modifying connections:
Security notice: SSH is only used to establish the initial connection to a target device, perform checks, and upload necessary files. The actual profiling commands and data are transferred through a raw, unencrypted socket. Nsight Systems should not be used in a network setup where attacker-in-the-middle attack is possible, or where untrusted parties may have network access to the target device.
While connecting to the target device, you will be prompted to input the user’s password. Please note that if you choose to remember the password, it will be stored in plain text in the configuration file on the host. Stored passwords are bound to the public key fingerprint of the remote device.
The No authentication option is useful for devices configured for passwordless login using root username. To enable such a configuration, edit the file /etc/ssh/sshd_config on the target and specify the following option:
PermitRootLogin yes
Then set empty password using passwd and restart the SSH service with service ssh restart.
Open ports: The Nsight Systems daemon requires port 22 and port 45555 to be open for listening. You can confirm that these ports are open with the following command:
sudo firewall-cmd --list-ports --permanent
sudo firewall-cmd --reload
To open a port use the following command, skip --permanent option to open only for this session:
sudo firewall-cmd --permanent --add-port 45555/tcp
sudo firewall-cmd --reload
Likewise, if you are running on a cloud system, you must open port 22 and port 45555 for ingress.
Kernel Version Number - To check for the version number of the kernel support of Nsight Systems on a target device, run the following command on the remote device:
cat /proc/quadd/version
Minimal supported version is 1.82.
Additionally, presence of Netcat command (nc) is required on the target device. For example, on Ubuntu this package can be installed using the following command:
sudo apt-get install netcat-openbsd
System-Wide Profiling Options
Target Sampling Options
Target sampling behavior is somewhat different for Nsight Systems Workstation Edition and Nsight Systems Embedded Platforms Edition.
Hotkey Trace Start/Stop
Nsight Systems Workstation Edition can use hotkeys to control profiling. Press the hotkey to start and/or stop a trace session from within the target application’s graphic window. This is useful when tracing games and graphic applications that use fullscreen display. In these scenarios switching to Nsight Systems’ UI would unnecessarily introduce the window manager’s footprint into the trace. To enable the use of Hotkey check the Hotkey checkbox in the project settings page:
The default hotkey is F12.
Launching Processes
Nsight Systems can launch new processes for profiling on target devices. Profiler ensures that all environment variables are set correctly to successfully collect trace information
The Edit arguments… link will open an editor window, where every command line argument is edited on a separate line. This is convenient when arguments contain spaces or quotes.
Profiling Windows Targets from the GUI
Profiling on Windows devices is similar to the profiling on Linux devices. Please refer to the Profiling Linux Targets from the GUI section for the detailed documentation and connection information. The major differences on the platforms are listed below:
Remoting to a Windows Based Machine
To perform remote profiling to a target Windows based machines, install and configure an OpenSSH Server on the target machine.
Hotkey Trace Start/Stop
Nsight Systems Workstation Edition can use hotkeys to control profiling. Press the hotkey to start and/or stop a trace session from within the target application’s graphic window. This is useful when tracing games and graphic applications that use fullscreen display. In these scenarios switching to Nsight Systems’ UI would unnecessarily introduce the window manager’s footprint into the trace. To enable the use of Hotkey check the Hotkey checkbox in the project settings page:
The default hotkey is F12.
Changing the Default Hotkey Binding - A different hotkey binding can be configured by setting the HotKeyIntValue configuration field in the config.ini file.
Set the decimal numeric identifier of the hotkey you would like to use for triggering start/stop from the target app graphics window. The default value is 123 which corresponds to 0x7B, or the F12 key.
Virtual key identifiers are detailed in MSDN’s Virtual-Key Codes.
Note that you must convert the hexadecimal values detailed in this page to their decimal counterpart before using them in the file. For example, to use the F1 key as a start/stop trace hotkey, use the following settings in the config.ini file:
HotKeyIntValue=112
Target Sampling Options on Windows
Nsight Systems can sample one process tree. Sampling here means interrupting each processor periodically. The sampling rate is defined in the project settings and is either 100Hz, 1KHz (default value), 2Khz, 4KHz, or 8KHz.
On Windows, Nsight Systems can collect thread activity of one process tree. Collecting thread activity means that each thread context switch event is logged and (optionally) a backtrace is collected at the point that the thread is scheduled back for execution. Thread states are displayed on the timeline.
If it was collected, the thread backtrace is displayed when hovering over a region where the thread execution is blocked.
Symbol Locations
Symbol resolution happens on host, and therefore does not affect performance of profiling on the target.
Press the Symbol locations… button to open the Configure debug symbols location dialog.
Use this dialog to specify:
Paths of PDB files
Symbols servers
The location of the local symbol cache
To use a symbol server:
Install Debugging Tools for Windows, a part of the Windows 10 SDK.
Add the symbol server URL using the Add Server button.
Information about Microsoft’s public symbol server, which enables getting Windows operating system related debug symbols can be found here.
Profiling QNX Targets from the GUI
Profiling on QNX devices is similar to the profiling on Linux devices. Please refer to the Profiling Linux Targets from the GUI section for the detailed documentation. The major differences on the platforms are listed below:
Backtrace sampling is not supported. Instead backtraces are collected for long OS runtime libraries calls. Please refer to the OS Runtime Libraries Trace section for the detailed documentation.
CUDA support is limited to CUDA 9.0+
Filesystem on QNX device might be mounted read-only. In that case Nsight Systems is not able to install target-side binaries, required to run the profiling session. Please make sure that target filesystem is writable before connecting to QNX target. For example, make sure the following command works:
echo XX > /xx && ls -l /xx
Profiling within JupyterLab
The JupyterLab Nsight extension integrates Nsight Systems profiling into jupyterLab for profiling of Jupyter notebook cells. CUDA kernels launched by the cells as well as CUDA and Python code execution can be profiled and analyzed.
For more information and to install the extension, go to JupyterLab Nsight extension on PyPI
Container and Scheduler Support
Collecting Data Within a Container
While examples in this section use Docker container semantics, other containers work much the same.
The following information assumes the reader is knowledgeable regarding Docker containers. For further information about Docker use in general, see the Docker documentation.
We strongly recommend using the CLI to profile in a container. Best container practice is to split services across containers when they do not require colocation. The Nsight Systems GUI is not needed to profile and brings in many dependencies, so the CLI is recommended. If you wish the GUI can be in a separate side-car container you use after to view your report. All you need is a shared folder between the containers. See section on GUI VNC container for more information.
Enable Docker Collection
When starting the Docker to perform a Nsight Systems collection, additional
steps are required to enable the perf_event_open system call. This is
required in order to utilize the Linux kernel’s perf subsystem which provides
sampling information to Nsight Systems.
There are three ways to enable the perf_event_open syscall. You can enable
it by using the --privileged=true switch, adding --cap-add=SYS_ADMIN
switch to your docker run command file, or you can enable it by setting the
seccomp security profile if your system meets the requirements.
Secure computing mode (seccomp) is a feature of the Linux kernel that can be used to restrict an application’s access. This feature is available only if the kernel is enabled with seccomp support. To check for seccomp support:
$ grep CONFIG_SECCOMP= /boot/config-$(uname -r)
The official Docker documentation says:
"Seccomp profiles require seccomp 2.2.1 which is not available on Ubuntu
14.04, Debian Wheezy, or Debian Jessie. To use seccomp on these distributions,
you must download the latest static Linux binaries (rather than packages)."
Download the default seccomp profile file, default.json, relevant to your Docker
version. If perf_event_open is already listed in the file as guarded by
CAP_SYS_ADMIN, then remove the perf_event_open line. Add the following
lines under “syscalls” and save the resulting file as default_with_perf.json.
{
"name": "perf_event_open",
"action": "SCMP_ACT_ALLOW",
"args": []
},
Then you will be able to use the following switch when starting the Docker to apply the new seccomp profile.
--security-opt seccomp=default_with_perf.json
Launch Docker Collection
Here is an example command that has been used to launch a Docker for testing with Nsight Systems:
sudo nvidia-docker run --network=host --security-opt seccomp=default_with_perf.json --rm -ti caffe-demo2 bash
There is a known issue where Docker collections terminate prematurely with older versions of the driver and the CUDA Toolkit. If collection is ending unexpectedly, please update to the latest versions.
After the Docker has been started, use the Nsight Systems CLI to launch a collection within the Docker. The resulting file can be imported into the Nsight Systems host like any other CLI result.
Profiling Services Launched via Kubernetes
Nsight Systems now can provide profiling via sidecar injection without need to modify your containers or k8/helm specs.
Once the sidecar is enabled, the data collected data can be filtered by namespace or pod using Kubernetes labels, or within a container process by using command-line regex.
This functionality is compatible with various cloud service provider’s in-house managed kubernetes variants including AKS, EKS, GKE, and OKE.
Documentation and download for this sidecar is available at NCG Sidecar Injector.
GUI VNC container
Nsight Systems provides a build script to build a self isolated Docker container with the Nsight Systems GUI and VNC server.
You can find the build.py script in the host-linux-x64/Scripts/VncContainer
directory (or similar on other architectures) under your Nsight Systems
installation directory. You will need to have
Docker, and Python 3.5 or later.
Available Parameters
Short Name |
Full Name |
Description |
|---|---|---|
|
(optional) Default password for VNC access (at least 6 characters). If it is specified and empty user will be asked during the build. Can be changed when running a container. |
|
-aba |
|
(optional) Additional arguments, which will be passed to the
|
-hd |
|
(optional) The directory with Nsight Systems host binaries (with GUI). |
-td |
|
(optional, repeatable) The directory with Nsight Systems target binaries (can be specified multiple times). |
|
(optional) Use TigerVNC instead of x11vnc. |
|
|
(optional) Install noVNC in the Docker container for HTTP access. |
|
|
(optional) Install xRDP in the Docker for RDP access. |
|
|
(optional) Default VNC server resolution in the format WidthxHeight (default 1920x1080). |
|
|
(optional) The directory to save temporary files (with the write access for the current user). By default, script or tmp directory will be used. |
Ports
These ports can be published from the container to provide access to the Docker container:
Port |
Purpose |
Condition |
|---|---|---|
TCP 5900 |
Port for VNC access |
|
TCP 80 (optional) |
Port for HTTP access to noVNC server |
Container is build with |
TCP 3389 (optional) |
Port for RDP access |
Container is build with |
Volumes
Docker folder |
Purpose |
Description |
|---|---|---|
/mnt/host |
Root path for shared folders |
Folder owned by the Docker user (inner content can be accessed from Nsight Systems GUI) |
/mnt/host/Projects |
Folder with projects and reports, created by Nsight Systems UI in container |
|
/mnt/host/logs |
Folder with inner services logs |
May be useful to send reports to developers |
Environment variables
Variable Name |
Purpose |
|---|---|
VNC_PASSWORD |
Password for VNC access (at least 6 characters) |
NSYS_WINDOW_WIDTH |
Width of VNC server display (in pixels) |
NSYS_WINDOW_HEIGHT |
Height of VNC server display (in pixels) |
Examples
With VNC access on port 5916:
sudo docker run -p 5916:5900/tcp -ti nsys-ui-vnc:1.0
With VNC access on port 5916 and HTTP access on port 8080:
sudo docker run -p 5916:5900/tcp -p 8080:80/tcp -ti nsys-ui-vnc:1.0
With VNC access on port 5916, HTTP access on port 8080 and RDP access on port 33890:
sudo docker run -p 5916:5900/tcp -p 8080:80/tcp -p 33890:3389/tcp -ti nsys-ui-vnc:1.0
With VNC access on port 5916, shared “HOME” folder from the host, VNC server resolution 3840x2160, and custom VNC password
sudo docker run -p 5916:5900/tcp -v $HOME:/mnt/host/home -e NSYS_WINDOW_WIDTH=3840 -e NSYS_WINDOW_HEIGHT=2160 -e VNC_PASSWORD=7654321 -ti nsys-ui-vnc:1.0
With VNC access on port 5916, shared “HOME” folder from the host, and the projects folder to access reports created by Nsight Systems GUI in container
sudo docker run -p 5916:5900/tcp -v $HOME:/mnt/host/home -v /opt/NsysProjects:/mnt/host/Projects -ti nsys-ui-vnc:1.0
GUI WebRTC container
Instructions for creating a self-isolated Docker container for accessing Nsight Systems through browser using WebRTC.
Prerequisites
x86_64 Linux
Internet access for downloading Ubuntu packages inside the container.
Build
To build the docker container use the follwing command:
$ sudo ./build.sh
The above command will create a docker image, which can be run using run.sh
Additional docker build arguments
Additional Docker Build arguments may be passed to the build.sh. For example:
$ sudo ./build.sh --network=host
Run
To run the docker container:
$ sudo ./run.sh
At the end of run.sh it will provide you with a URL to connect to the WebRTC client. It will look something like http://$HOST_IP:8080/. You can use this address in your browser to access Nsight Systems GUI interface.
Additional docker run arguments
Additional Docker Run arguments may be passed to the run.sh. These argument can be used to mount host directories with Nsight Systems reports to the docker container. For example:
$ sudo ./run.sh -v $HOME:/mnt/host/home -v /myawesomereports:/mnt/host/myawesomereports
Runtime environment variables
Runtime environment variables can be used to configure runtime parameters.
Variable |
Description |
Default Value |
|---|---|---|
HOST_IP |
IP of the server that will be sent to client. This IP should be accessible from the client side to establish client/server connection. |
The IP address of the first available network interface. |
HTTP_PORT |
TCP port for HTTP access to Nsight Systems user interface. |
8080 |
MEDIA_PORT |
TCP port that will be used for WebRTC data transmission. |
3478 |
SCREEN |
Resolution of the screen used for rendering. Only 1920x1080, 1280x720, 1152x648, 1024x576, 960x720, 800x600 are currently supported |
1920x1080 |
CONTAINER_NAME |
Name which will be assigned to a running container. |
nvidia-devtools-streamer-nsys |
Volumes
List of usefull internal Docker folders:
Docker folder |
Purpose |
Description |
|---|---|---|
/mnt/host/logs |
Folder with inner services logs |
May be useful to send reports to NVIDIA developer |
Example
To run the container on 10.10.10.10 network interface, using 8000 HTTP port, 8888 media port:
$ sudo HOST_IP=10.10.10.10 HTTP_PORT=8000 MEDIA_PORT=3479 ./run.sh
Stop
To stop the docker container the command below can be used:
$ sudo ./stop.sh
If the CONTAINER_NAME` environment variable was used to specify the name of a container during its start-up, the same variable should also be used when issuing the command to stop the container.
Migrating from NVIDIA nvprof
Using the Nsight Systems CLI nvprof Command
The nvprof command of the Nsight Systems CLI is intended to help former nvprof users transition to nsys. Many nvprof switches are not supported by nsys, often because they are now part of NVIDIA Nsight Compute.
The full nvprof documentation can be found at https://docs.nvidia.com/cuda/profiler-users-guide.
The nvprof transition guide for Nsight Compute can be found at https://docs.nvidia.com/nsight-compute/NsightComputeCli/index.html#nvprof-guide.
Any nvprof switch not listed below is not supported by the nsys nvprof command. No additional nsys functionality is available through this command. New features will not be added to this command in the future.
CLI nvprof Command Switch Options
After choosing the nvprof command switch, the following options are available. When you are ready to move to using Nsight Systems CLI directly, see Command Line Options documentation for the nsys switch(es) given below. Note that the nsys implementation and output may vary from nvprof.
Usage.
nsys nvprof [options]
Switch |
Parameters (Default in Bold) |
nsys switch |
Switch Description |
|---|---|---|---|
|
off, openmpi, mpich |
|
Automatically annotate MPI calls with NVTX markers. Specify the MPI implementation installed on your machine. Only OpenMPI and MPICH implementations are supported. |
|
on, off |
|
Collect information about CPU thread API activity. |
|
none, runtime, driver,all |
|
Turn on/off CUDA runtime and driver API tracing. For Nsight Systems there is no separate CUDA runtime and CUDA driver trace, so selecting |
|
on, off |
if off use |
Enable/disable profiling from the start of the application. If disabled, the application can use {cu,cuda}Profiler{Start,Stop} to turn on/off profiling. |
|
<nanoseconds> default=0 |
|
If greater than 0, stop the collection and kill the launched application after timeout seconds. nvprof started counting when the CUDA driver is initialized. nsys starts counting immediately. |
|
on, off |
|
Turn on/off CPU profiling |
|
on, off |
|
Enable/disable recording information from the OpenACC profiling interface. Note: OpenACC profiling interface depends on the presence of the OpenACC runtime. For supported runtimes, see CUDA Trace section of documentation |
|
<filename> |
|
Export named file to be imported or opened in the Nsight Systems GUI. %q{ENV_VAR} in string will be replaced with the set value of the environment variable. If not set this is an error. %h in the string is replaced with the system hostname. %% in the string is replaced with %. %p in the string is not supported currently. Any other character following % is illegal. The default is report1, with the number incrementing to avoid overwriting files, in users working directory. |
|
|
Force overwriting all output files with same name. |
|
|
|
Print Nsight Systems CLI help |
|
|
|
Print Nsight Systems CLI version information |
Next Steps
NVIDIA Visual Profiler (NVVP) and NVIDIA nvprof are deprecated. New GPUs and features will not be supported by those tools. We encourage you to make the move to Nsight Systems now. For additional information, suggestions, and rationale, see the blog series in Other Resources.
Direct3D Trace
Nsight Systems has the ability to trace both the Direct3D 11 API and the Direct3D 12 API on Windows targets.
D3D11 API trace
Nsight Systems can capture information about Direct3D 11 API calls made by the profiled process. This includes capturing the execution time of D3D11 API functions, performance markers, and frame durations.
SLI Trace
Trace SLI queries and peer-to-peer transfers of D3D11 applications. Requires SLI hardware and an active SLI profile definition in the NVIDIA console.
D3D12 API Trace
Direct3D 12 is a low-overhead 3D graphics and compute API for Microsoft Windows. Information about Direct3D 12 can be found at the Direct3D 12 Programming Guide.
Nsight Systems can capture information about Direct3D 12 usage by the profiled process. This includes capturing the execution time of D3D12 API functions, corresponding workloads executed on the GPU, performance markers, and frame durations.
The Command List Creation row displays time periods when command lists were being created. This enables developers to improve their application’s multi-threaded command list creation. Command list creation time period is measured between the call to ID3D12GraphicsCommandList::Reset and the call to ID3D12GraphicsCommandList::Close.
The GPU row shows a compressed view of the D3D12 queue activity, color-coded by the queue type. Expanding it will show the individual queues and their corresponding API calls.
A Command Queue row is displayed for each D3D12 command queue created by the profiled application. The row’s header displays the queue’s running index and its type (Direct, Compute, Copy).
The DX12 API Memory Ops row displays all API memory operations and non-persistent resource mappings. Event ranges in the row are color-coded by the heap type they belong to (Default, Readback, Upload, Custom, or CPU-Visible VRAM), with usage warnings highlighted in yellow. A breakdown of the operations can be found by expanding the row to show rows for each individual heap type.
The following operations and warnings are shown:
Calls to
ID3D12Device::CreateCommittedResource,ID3D12Device4::CreateCommittedResource1, andID3D12Device8::CreateCommittedResource2A warning will be reported if
D3D12_HEAP_FLAG_CREATE_NOT_ZEROEDis not set in the method’sHeapFlagsparameter
Calls to
ID3D12Device::CreateHeapandID3D12Device4::CreateHeap1A warning will be reported if
D3D12_HEAP_FLAG_CREATE_NOT_ZEROEDis not set in theFlagsfield of the method’spDescparameter
Calls to
ID3D12Resource::ReadFromSubResourceA warning will be reported if the read is to a
D3D12_CPU_PAGE_PROPERTY_WRITE_COMBINECPU page or from aD3D12_HEAP_TYPE_UPLOADresource
Calls to
ID3D12Resource::WriteToSubResourceA warning will be reported if the write is from a
D3D12_CPU_PAGE_PROPERTY_WRITE_BACKCPU page or to aD3D12_HEAP_TYPE_READBACKresource
Calls to
ID3D12Resource::MapandID3D12Resource::Unmapwill be matched into [Map, Unmap] ranges for non-persistent mappings. If a mapping range is nested, only the most external range (reference count = 1) will be shown.
The API row displays time periods where ID3D12CommandQueue::ExecuteCommandLists was called. The GPU Workload row displays time periods where workloads were executed by the GPU. The workload’s type (Graphics, Compute, Copy, etc.) is displayed on the bar representing the workload’s GPU execution.
In addition, you can see the PIX command queue CPU-side performance markers, GPU-side performance markers and the GPU Command List performance markers, each in their row.
Clicking on a GPU workload highlights the corresponding ID3D12CommandQueue::ExecuteCommandLists, ID3D12GraphicsCommandList::Reset and ID3D12GraphicsCommandList::Close API calls, and vice versa.
Detecting which CPU thread was blocked by a fence can be difficult in complex apps that run tens of CPU threads. The timeline view displays the 3 operations involved:
The CPU thread pushing a signal command and fence value into the command queue. This is displayed on the DX12 Synchronization sub-row of the calling thread.
The GPU executing that command, setting the fence value and signaling the fence. This is displayed on the GPU Queue Synchronization sub-row.
The CPU thread calling a Win32 wait API to block-wait until the fence is signaled. This is displayed on the Thread’s OS runtime libraries row.
Clicking one of these will highlight it and the corresponding other two calls.
Nsight Systems D3D12 trace captures D3D12 Work Graphs dispatch calls to DispatchGraph and time boundaries of the GPU execution of the work graph.
The DX12 API row displays ID3D12GraphicsCommandList10::DispatchGraph calls. The GPU PIX Markers row marks graph execution by the GPU with a custom marker captioned “D3D12 graph execution”.
WDDM Queues
The Windows Display Driver Model (WDDM) architecture uses queues to send work packets from the CPU to the GPU. Each D3D device in each process is associated with one or more contexts. Graphics, compute, and copy commands that the profiled application uses are associated with a context, batched in a command buffer, and pushed into the relevant queue associated with that context.
Nsight Systems can capture the state of these queues during the trace session.
Enabling the “Collect additional range of ETW events” option will also capture extended DxgKrnl events from the Microsoft-Windows-DxgKrnl provider, such as context status, allocations, sync wait, signal events, etc.
A command buffer in a WDDM queues may have one the following types:
Render
Deferred
System
MMIOFlip
Wait
Signal
Device
Software
It may also be marked as a Present buffer, indicating that the application has finished rendering and requests to display the source surface.
See the Microsoft documentation for the WDDM architecture and the DXGKETW_QUEUE_PACKET_TYPE enumeration.
To retain the .etl trace files captured, so that they can be viewed in other tools (e.g. GPUView), change the “Save ETW log files in project folder” option under “Profile Behavior” in Nsight Systems’s global Options dialog. The .etl files will appear in the same folder as the .nsys-rep file, accessible by right-clicking the report in the Project Explorer and choosing “Show in Folder…”. Data collected from each ETW provider will appear in its own .etl file, and an additional .etl file named “Report XX-Merged-*.etl”, containing the events from all captured sources, will be created as well.
WDDM HW Scheduler
When GPU Hardware Scheduling is enabled in Windows 10 or newer version, the Windows Display Driver Model (WDDM) uses the DxgKrnl ETW provider to expose report of NVIDIA GPUs’ hardware scheduling context switches.
Nsight Systems can capture these context switch events, and display under the GPUs in the timeline rows titled WDDM HW Scheduler - [HW Queue type]. The ranges under each queue will show the process name and PID assoicated with the GPU work during the time period.
The events will be captured if GPU Hardware Scheduling is enabled in the Windows System Display settings, and “Collect WDDM Trace” is enabled in the Nsight Systems Project Settings.
Vulkan API Trace
Vulkan Overview
Vulkan is a low-overhead, cross-platform 3D graphics and compute API, targeting a wide variety of devices from PCs to mobile phones and embedded platforms. The Vulkan API is defined by the Khronos Group. Information about Vulkan and the Khronos Group can be found at the Khronos Vulkan Site.
Nsight Systems can capture information about Vulkan usage by the profiled process. This includes capturing the execution time of Vulkan API functions, corresponding GPU workloads, debug util labels, and frame durations. Vulkan profiling is supported on both Windows and x86 Linux operating systems.
The Command Buffer Creation row displays time periods when command buffers were being created. This enables developers to improve their application’s multi-threaded command buffer creation. Command buffer creation time period is measured between the call to vkBeginCommandBuffer and the call to vkEndCommandBuffer.
A Queue row is displayed for each Vulkan queue created by the profiled application. The API sub-row displays time periods where vkQueueSubmit was called. The GPU Workload sub-row displays time periods where workloads were executed by the GPU.
In addition, you can see Vulkan debug util labels on both the CPU and the GPU.
Clicking on a GPU workload highlights the corresponding vkQueueSubmit call, and vice versa.
The Vulkan Memory Operations row contains an aggregation of all the Vulkan host-side memory operations, such as host-blocking writes and reads or non-persistent map-unmap ranges.
The row is separated into sub-rows by heap index and memory type - the tooltip for each row and the ranges inside show the heap flags and the memory property flags.
Pipeline Creation Feedback
When tracing target application calls to Vulkan pipeline creation APIs, Nsight Systems leverages the Pipeline Creation Feedback extension to collect more details about the duration of individual pipeline creation stages.
See Pipeline Creation Feedback extension for details about this extension.
Vulkan pipeline creation feedback is available on NVIDIA driver release 435 or later.
Vulkan GPU Trace Notes
Vulkan GPU trace is available only when tracing apps that use NVIDIA GPUs.
The endings of Vulkan Command Buffers execution ranges on Compute and Transfer queues may appear earlier on the timeline than their actual occurrence.
Stutter Analysis
Stutter Analysis Overview
Nsight Systems on Windows targets displays stutter analysis visualization aids for profiled graphics applications that use either OpenGL, D3D11, D3D12 or Vulkan, as detailed below in the following sections.
FPS Overview
The Frame Duration section displays frame durations on both the CPU and the GPU.
The frame duration row displays live FPS statistics for the current timeline viewport. Values shown are:
Number of CPU frames shown of the total number captured
Average, minimal, and maximal CPU frame time of the currently displayed time range
Average FPS value for the currently displayed frames
The 99th percentile value of the frame lengths (such that only 1% of the frames in the range are longer than this value).
The values will update automatically when scrolling, zooming or filtering the timeline view.
The stutter row highlights frames that are significantly longer than the other frames in their immediate vicinity.
The stutter row uses an algorithm that compares the duration of each frame to the median duration of the surrounding 19 frames. Duration difference under 4 milliseconds is never considered a stutter, to avoid cluttering the display with frames whose absolute stutter is small and not noticeable to the user.
For example, if the stutter threshold is set at 20%:
Median duration is 10 ms. Frame with 13 ms time will not be reported (relative difference > 20%, absolute difference < 4 ms)
Median duration is 60 ms. Frame with 71 ms time will not be reported (relative difference < 20%, absolute difference > 4 ms)
Median duration is 60 ms. Frame with 80 ms is a stutter (relative difference > 20%, absolute difference > 4 ms, both conditions met)
OSC detection
The “19 frame window median” algorithm by itself may not work well with some cases of “oscillation” (consecutive fast and slow frames), resulting in some false positives. The median duration is not meaningful in cases of oscillation and can be misleading.
To address the issue and identify if oscillating frames, the following method is applied:
For every frame, calculate the median duration, 1st and 3rd quartiles of 19-frames window.
Calculate the delta and ratio between 1st and 3rd quartiles.
If the 90th percentile of 3rd - 1st quartile delta array > 4 ms AND the 90th percentile of 3rd/1st quartile array > 1.2 (120%) then mark the results with “OSC” text.
Right-clicking the Frame Duration row caption lets you choose the target frame rate (30, 60, 90 or custom frames per second).
By clicking the Customize FPS Display option, a customization dialog pops up. In the dialog, you can now define the frame duration threshold to customize the view of the potentially problematic frames. In addition, you can define the threshold for the stutter analysis frames.
Frame duration bars are color coded:
Green, the frame duration is shorter than required by the target FPS ratio.
Yellow, duration is slightly longer than required by the target FPS rate.
Red, duration far exceeds that required to maintain the target FPS rate.
The CPU Frame Duration row displays the CPU frame duration measured between the ends of consecutive frame boundary calls:
The OpenGL frame boundaries are
eglSwapBuffers/glXSwapBuffers/SwapBufferscalls.The D3D11 and D3D12 frame boundaries are
IDXGISwapChainX::Presentcalls.The Vulkan frame boundaries are
vkQueuePresentKHRcalls.
The timing of the actual calls to the frame boundary calls can be seen in the blue bar at the bottom of the CPU frame duration row
The GPU Frame Duration row displays the time measured between
The start time of the first GPU workload execution of this frame.
The start time of the first GPU workload execution of the next frame.
Reflex SDK
NVIDIA Reflex SDK is a series of NVAPI calls that allow applications to integrate the Ultra Low Latency driver feature more directly into their game to further optimize synchronization between simulation and rendering stages and lower the latency between user input and final image rendering. For more details about Reflex SDK, see Reflex SDK Site.
Nsight Systems will automatically capture NVAPI functions when either Direct3D 11, Direct3D 12, or Vulkan API trace are enabled.
The Reflex SDK row displays timeline ranges for the following types of latency markers:
RenderSubmit.
Simulation.
Present.
Driver.
OS Render Queue.
GPU Render.
Performance Warnings row
This row shows performance warnings and common pitfalls that are automatically detected based on the enabled capture types. Warnings are reported for:
ETW performance warnings
Vulkan calls to
vkQueueSubmitand D3D12 calls toID3D12CommandQueue::ExecuteCommandListthat take a longer time to execute than the total time of the GPU workloads they generatedUsage of Vulkan API functions that may adversely affect performance
Creation of a Vulkan device with memory zeroing, whether by physical device default or manually
Vulkan command buffer barrier which can be combined or removed, such as subsequent barriers or read-to-read barriers
Frame Health
The Frame Health row displays actions that took significantly a longer time during the current frame, compared to the median time of the same actions executed during the surrounding 19-frames. This is a great tool for detecting the reason for frame time stuttering. Such actions may be: shader compilation, present, memory mapping, and more. Nsight Systems measures the accumulated time of such actions in each frame. For example: calculating the accumulated time of shader compilations in each frame and comparing it to the accumulated time of shader compilations in the surrounding 19 frames.
Example of a Vulkan frame health row:
Windows GPU Memory Utilization
Each GPU has two rows detailing its memory utilization: GPU VRAM, showing the memory consumed on the device, and GPU WDDM SYSMEM, showing the memory consumed on the host computer RAM.
These rows show a green-colored line graph for the memory budget for this memory segment, and an orange-colored line graph for the actual amount of memory used. Note that these graphs are scaled to fit the highest value enconutered, as indicated by the “Y axis” value in the row header. You can use the vertical zoom slider in the top-right of the timeline view to make the row taller and view the graph in more detail.
Note that the value in the GPU VRAM row is not the same as the CUDA kernel memory allocation graph, see CUDA GPU Memory Allocation Graph for that functionality.
The GPU VRAM row also has several child rows, accessed by expanding the row in the tree view
VidMm Device Suspension
This row displays time ranges when the GPU memory manager suspended all memory transfer operations, pending the completion of a single memory transfer.
Demoted Memory
This row displays the amount of VRAM that was demoted from GPU local memory to non-local memory (possibly due to exceeding the VRAM budget) as a blue-colored line graph. High amounts of demoted memory could be indicative of video memory leaks or poor memory management. Note that the Demoted memory row is scaled to its highest value, similar to the GPU VRAM and GPU WDDM SYSMEM rows.
Resource Allocations
This row shows markers indicating resource allocation events. VRAM resources are shown as green markers while SYSMEM resources are shown in gray. Hovering over a marker or selecting it in the Events view will display all the allocation parameters as well as the call stack that led to the allocation event.
Resource Migrations
This row displays a breakdown of resources’ movement between VRAM and SYSMEM, focusing on resource evictions. The main row shows a timeline of total evicted resource memory and count as a red-colored line graph.
Each child row displays a timeline of the status of each resource, as reflected
by WDDM events related to it. If the object has been named using PIX or
ID3D11Object::SetName / ID3D12Object::SetName, the name will be shown
in the row title. Whether named or not, the row title will also show the
resource dimensions, format, priority, and the memory size migrated. If the
resource was migrated in parts using subresources, the row can be expanded to
show the status for each subresource at any given time.
Expanding the row for a resource will show the individual WDDM events relevant to it and the call stacks that led to each event.
By default, the resources are sorted by Relevance (most / largest migrations). Right-clicking the main Resource Migrations row header allows choosing between the following sorting options:
Relevance
Name
Format
Priority
Earliest allocation timestamp (order of appearance on the host)
Earliest migration timestamp (order of appearance on the device)
The top 5 resources are shown initially. If more than 5 resources exist, a row showing the number of hidden resources and buttons allowing to show more or fewer of them will appear below them. Right-click this row and select “show all” or “show all collapsed” to display all the resources at once.
Memory Transfer
This row shows an overview of all memory transfer operations. Device-to-host transfers are shown in orange, host-to-device transfers are shown in green, discarded device memory is shown in light green, and unknown events are shown in dark gray. The height of each event marker corresponds to the amount of memory that the event affected. Hovering over the marker will show the exact amount.
Expanding the row will show a breakdown of the events by each specific type.
Vertical Synchronization
The VSYNC rows display when the monitor’s vertical synchronizations occur.
OpenMP Trace
Nsight Systems for Linux is capable of capturing information about OpenMP events. This functionality is built on the OpenMP Tools Interface (OMPT), full support is available only for runtime libraries supporting tools interface defined in OpenMP 5.0 or greater.
As an example, LLVM OpenMP runtime library partially implements tools interface. If you use PGI compiler <= 20.4 to build your OpenMP applications, add -mp=libomp switch to use LLVM OpenMP runtime and enable OMPT based tracing. If you use Clang, make sure the LLVM OpenMP runtime library you link to was compiled with tools interface enabled.
Only a subset of the OMPT callbacks are processed:
ompt_callback_parallel_begin
ompt_callback_parallel_end
ompt_callback_sync_region
ompt_callback_task_create
ompt_callback_task_schedule
ompt_callback_implicit_task
ompt_callback_master
ompt_callback_reduction
ompt_callback_task_create
ompt_callback_cancel
ompt_callback_mutex_acquire, ompt_callback_mutex_acquired
ompt_callback_mutex_acquired, ompt_callback_mutex_released
ompt_callback_mutex_released
ompt_callback_work
ompt_callback_dispatch
ompt_callback_flush
Note
The raw OMPT events are used to generate ranges indicating the runtime of OpenMP operations and constructs.
Example screenshot:
OS Runtime Libraries Trace
On Linux, OS runtime libraries can be traced to gather information about low-level userspace APIs. This traces the system call wrappers and thread synchronization interfaces exposed by the C runtime and POSIX Threads (pthread) libraries. This does not perform a complete runtime library API trace, but instead focuses on the functions that can take a long time to execute, or could potentially cause your thread be unscheduled from the CPU while waiting for an event to complete. OS runtime trace is not available for Windows targets.
OS runtime tracing complements and enhances sampling information by:
Visualizing when the process is communicating with the hardware, controlling resources, performing multi-threading synchronization or interacting with the kernel scheduler.
Adding additional thread states by correlating how OS runtime libraries traces affect the thread scheduling:
Waiting — the thread is not scheduled on a CPU, it is inside of an OS runtime libraries trace and is believed to be waiting on the firmware to complete a request.
In OS runtime library function — the thread is scheduled on a CPU and inside of an OS runtime libraries trace. If the trace represents a system call, the process is likely running in kernel mode.
Collecting backtraces for long OS runtime libraries call. This provides a way to gather blocked-state backtraces, allowing you to gain more context about why the thread was blocked so long, yet avoiding unnecessary overhead for short events.
To enable OS runtime libraries tracing from Nsight Systems:
CLI — Use the -t, --trace option with the osrt parameter. See Command Line Options for more information.
GUI — Select the Collect OS runtime libraries trace checkbox.
You can also use Skip if shorter than. This will skip calls shorter than the given threshold. Enabling this option will improve performances as well as reduce noise on the timeline. We strongly encourage you to skip OS runtime libraries call shorter than 1 μs.
Locking a Resource
The functions listed below receive a special treatment. If the tool detects that the resource is already acquired by another thread and will induce a blocking call, we always trace it. Otherwise, it will never be traced.
pthread_mutex_lock
pthread_rwlock_rdlock
pthread_rwlock_wrlock
pthread_spin_lock
sem_wait
Note that even if a call is determined as potentially blocking, there is a chance that it may not actually block after a few cycles have elapsed. The call will still be traced in this scenario.
Limitations
Nsight Systems only traces syscall wrappers exposed by the C runtime. It is not able to trace syscall invoked through assembly code.
Additional thread states, as well as backtrace collection on long calls, are only enabled if sampling is turned on.
It is not possible to configure the depth and duration threshold when collecting backtraces. Currently, only OS runtime libraries calls longer than 80 μs will generate a backtrace with a maximum of 24 frames. This limitation will be removed in a future version of the product.
It is required to compile your application and libraries with the
-funwind-tablescompiler flag in order for Nsight Systems to unwind the backtraces correctly.
OS Runtime Libraries Trace Filters
The OS runtime libraries tracing is limited to a select list of functions. It also depends on the version of the C runtime linked to the application.
OS Runtime Default Function List
Libc system call wrappers
accept
accept4
acct
alarm
arch_prctl
bind
bpf
brk
chroot
clock_nanosleep
connect
copy_file_range
creat
creat64
dup
dup2
dup3
epoll_ctl
epoll_pwait
epoll_wait
fallocate
fallocate64
fcntl
fdatasync
flock
fork
fsync
ftruncate
futex
ioctl
ioperm
iopl
kill
killpg
listen
membarrier
mlock
mlock2
mlockall
mmap
mmap64
mount
move_pages
mprotect
mq_notify
mq_open
mq_receive
mq_send
mq_timedreceive
mq_timedsend
mremap
msgctl
msgget
msgrcv
msgsnd
msync
munmap
nanosleep
nfsservctl
open
open64
openat
openat64
pause
pipe
pipe2
pivot_root
poll
ppoll
prctl
pread
pread64
preadv
preadv2
preadv64
process_vm_readv
process_vm_writev
pselect6
ptrace
pwrite
pwrite64
pwritev
pwritev2
pwritev64
read
readv
reboot
recv
recvfrom
recvmmsg
recvmsg
rt_sigaction
rt_sigqueueinfo
rt_sigsuspend
rt_sigtimedwait
sched_yield
seccomp
select
semctl
semget
semop
semtimedop
send
sendfile
sendfile64
sendmmsg
sendmsg
sendto
shmat
shmctl
shmdt
shmget
shutdown
sigaction
sigsuspend
sigtimedwait
socket
socketpair
splice
swapoff
swapon
sync
sync_file_range
syncfs
tee
tgkill
tgsigqueueinfo
tkill
truncate
umount2
unshare
uselib
vfork
vhangup
vmsplice
wait
wait3
wait4
waitid
waitpid
write
writev
_sysctl
POSIX Threads
pthread_barrier_wait
pthread_cancel
pthread_cond_broadcast
pthread_cond_signal
pthread_cond_timedwait
pthread_cond_wait
pthread_create
pthread_join
pthread_kill
pthread_mutex_lock
pthread_mutex_timedlock
pthread_mutex_trylock
pthread_rwlock_rdlock
pthread_rwlock_timedrdlock
pthread_rwlock_timedwrlock
pthread_rwlock_tryrdlock
pthread_rwlock_trywrlock
pthread_rwlock_wrlock
pthread_spin_lock
pthread_spin_trylock
pthread_timedjoin_np
pthread_tryjoin_np
pthread_yield
sem_timedwait
sem_trywait
sem_wait
I/O
aio_fsync
aio_fsync64
aio_suspend
aio_suspend64
fclose
fcloseall
fflush
fflush_unlocked
fgetc
fgetc_unlocked
fgets
fgets_unlocked
fgetwc
fgetwc_unlocked
fgetws
fgetws_unlocked
flockfile
fopen
fopen64
fputc
fputc_unlocked
fputs
fputs_unlocked
fputwc
fputwc_unlocked
fputws
fputws_unlocked
fread
fread_unlocked
freopen
freopen64
ftrylockfile
fwrite
fwrite_unlocked
getc
getc_unlocked
getdelim
getline
getw
getwc
getwc_unlocked
lockf
lockf64
mkfifo
mkfifoat
posix_fallocate
posix_fallocate64
putc
putc_unlocked
putwc
putwc_unlocked
Miscellaneous
forkpty
popen
posix_spawn
posix_spawnp
sigwait
sigwaitinfo
sleep
system
usleep
Syscall Trace (Preview)
Nsight Systems for Linux and Nsight Systems Embedded Platforms Edition are capable of tracing Linux system calls in kernel space. This feature uses Linux’s eBPF technology, and is supported on systems running Linux v5.11 or newer, specifically those that are built with CONFIG_DEBUG_INFO_BTF enabled, which is the default on most major Linux distros. This feature requires root privileges and can also be used inside Docker containers that are run with the --privileged option, or with CAP_SYS_ADMIN capabilities, or at a minimum, with CAP_BPF and CAP_PERFMON capabilities.
To enable this feature:
CLI — Set syscall as one of the arguments to the -t, --trace option.
GUI — Select the Collect syscall trace checkbox.
Example screenshot:
NVTX Trace
The NVIDIA Tools Extension Library (NVTX) is a powerful mechanism that allows users to manually instrument their application. Nsight Systems can then collect the information and present it on the timeline.
Nsight Systems supports version 3.0 of the NVTX specification.
The following features are supported:
Domains
nvtxDomainCreate(), nvtxDomainDestroy()
nvtxDomainRegisterString()
Push-pop ranges (nested ranges that start and end in the same thread).
nvtxRangePush(), nvtxRangePushEx()
nvtxRangePop()
nvtxDomainRangePushEx()
nvtxDomainRangePop()
Start-end ranges (ranges that are global to the process and are not restricted to a single thread)
nvtxRangeStart(), nvtxRangeStartEx()
nvtxRangeEnd()
nvtxDomainRangeStartEx()
nvtxDomainRangeEnd()
Marks
nvtxMark(), nvtxMarkEx()
nvtxDomainMarkEx()
Thread names
nvtxNameOsThread()
Categories
nvtxNameCategory()
nvtxDomainNameCategory()
To learn more about specific features of NVTX, please refer to the NVTX header
file: nvToolsExt.h or the NVTX documentation.
To use NVTX in your application, follow these steps:
Add
#include "nvtx3/nvToolsExt.h"in your source code. The nvtx3 directory is located in the Nsight Systems package in the Target-<architecture>/nvtx/include directory and is available via github at http://github.com/NVIDIA/NVTX.Add the following compiler flag:
-ldlAdd calls to the NVTX API functions. For example, try adding
nvtxRangePush("main")in the beginning of themain()function, andnvtxRangePop()just before the return statement in the end.For convenience in C++ code, consider adding a wrapper that implements RAII (resource acquisition is initialization) pattern, which would guarantee that every range gets closed.
In the project settings, select the Collect NVTX trace checkbox.
In addition, by enabling the “Insert NVTX Marker hotkey” option it is possible to add NVTX markers to a running non-console applications by pressing the F11 key. These will appear in the report under the NVTX Domain named “HotKey markers”.
Typically calls to NVTX functions can be left in the source code even if the application is not being built for profiling purposes, since the overhead is very low when the profiler is not attached.
NVTX is not intended to annotate very small pieces of code that are being called very frequently. A good rule of thumb to use: if code being annotated usually takes less than 1 microsecond to execute, adding an NVTX range around this code should be done carefully.
Note
Range annotations should be matched carefully. If many ranges are opened but not closed, Nsight Systems has no meaningful way to visualize it. A rule of thumb is to not have more than a couple dozen ranges open at any point in time. Nsight Systems does not support reports with many unclosed ranges.
NVTX Payloads and Counters (Preview)
NVTX Extended Payloads and NVTX Counters increase the flexibility of NVTX annotations by allowing users to pass arbitrary structured data to NVTX events. This then will allow users to define the layout of this data in the Nsight Systems UI without additional data conversion.
For more information, see NVTX documentation.
NVTX Domains and Categories
NVTX domains enable scoping of annotations. Unless specified differently, all events and annotations are in the default domain. Additionally, categories can be used to group events.
Nsight Systems gives the user the ability to include or exclude NVTX events from a particular domain. This can be especially useful if you are profiling across multiple libraries and are only interested in nvtx events from some of them.
This functionality is also available from the CLI. See the CLI documentation
for --nvtx-domain-include and --nvtx-domain-exclude for more details.
Categories that are set in by the user will be recognized and displayed in the GUI.
CUDA Trace
Basic CUDA trace
Nsight Systems is capable of capturing information about CUDA execution in the profiled process.
The following information can be collected and presented on the timeline in the report:
CUDA API trace — trace of CUDA Runtime and CUDA Driver calls made by the application.
CUDA Runtime calls typically start with
cudaprefix (e.g.cudaLaunch).CUDA Driver calls typically start with
cuprefix (e.g.cuDeviceGetCount).
CUDA workload trace — trace of activity happening on the GPU, which includes memory operations (e.g., Host-to-Device memory copies) and kernel executions. Within the threads that use the CUDA API, additional child rows will appear in the timeline tree.
On Nsight Systems Workstation Edition, cuDNN and cuBLAS API tracing and OpenACC tracing.
Near the bottom of the timeline row tree, the GPU node will appear and contain a CUDA node. Within the CUDA node, each CUDA context used within the process will be shown along with its corresponding CUDA streams. Steams will contain memory operations and kernel launches on the GPU. Kernel launches are represented by blue, while memory transfers are displayed in red.
The easiest way to capture CUDA information is to launch the process from Nsight Systems, and it will setup the environment for you. To do so, simply set up a normal launch and select the Collect CUDA trace checkbox.
For Nsight Systems Workstation Edition this looks like:
For Nsight Systems Embedded Platforms Edition this looks like:
Additional configuration parameters are available:
Collect backtraces for API calls longer than X seconds - turns on collection of CUDA API backtraces and sets the minimum time a CUDA API event must take before its backtraces are collected. Setting this value too low can cause high application overhead and seriously increase the size of your results file.
Flush data periodically — specifies the period after which an attempt to flush CUDA trace data will be made. Normally, in order to collect full CUDA trace, the application needs to finalize the device used for CUDA work (call
cudaDeviceReset(), and then let the application gracefully exit (as opposed to crashing).This option allows flushing CUDA trace data even before the device is finalized. However, it might introduce additional overhead to a random CUDA Driver or CUDA Runtime API call.
Skip some API calls — avoids tracing insignificant CUDA Runtime API calls (namely,
cudaConfigureCall(),cudaSetupArgument(),cudaHostGetDevicePointers()). Not tracing these functions allows Nsight Systems to significantly reduce the profiling overhead, without losing any interesting data.Collect GPU Memory Usage - collects information used to generate a graph of CUDA allocated memory across time. Note that this will increase overhead. See CUDA GPU Memory Allocation Graph.
Collect Unified Memory CPU page faults - collects information on page faults that occur when CPU code tries to access a memory page that resides on the device. See Unified Memory CPU Page Faults.
Collect Unified Memory GPU page faults - collects information on page faults that occur when GPU code tries to access a memory page that resides on the CPU. See Unified Memory GPU Page Faults.
Collect CUDA Graph trace - by default, CUDA tracing will collect and expose information on a whole graph basis. The user can opt to collect on a node per node basis. See CUDA Graph Trace.
For Nsight Systems Workstation Edition, Collect cuDNN trace, Collect cuBLAS trace, Collect OpenACC trace - selects which (if any) extra libraries that depend on CUDA to trace.
OpenACC versions 2.0, 2.5, and 2.6 are supported when using PGI runtime version 15.7 or greater and not compiling statically. In order to differentiate constructs, a PGI runtime of 16.1 or later is required. Note that Nsight Systems Workstation Edition does not support the GCC implementation of OpenACC at this time.
Note
If your application crashes before all collected CUDA trace data has been copied out, some or all data might be lost and not present in the report.
Note
Nsight Systems will not have information about CUDA events that were still in device buffers when analysis terminated. It is a good idea, if using cudaProfilerAPI to control analysis to call cudaDeviceReset before ending analysis.
Launching NVIDIA Nsight Compute from a CUDA Kernel
After you have used CUDA trace in Nsight Systems to locate a potential problem kernel, you may want to run NVIDIA Nsight Compute on that specific kernel. Right click on the kernel to bring up a menu.
If this is the first time that the user has selected this feature, then we show the following dialog box to get their preferences:
The first option invokes the NVIDIA Nsight Compute UI with known parameters. It is the preferred option for local or remote profiling. The user must provide the location of the ncu-ui executable. Nsight Systems will verify that the path and executable are valid.
The second option is provided for the convenience of users who do not have NVIDIA Nsight Compute installed on the host system, but simply want the command line they can use to run the Nsight Compute on the remote target by themselves without much automation.
If the user selects the option to start the NCU UI, Nsight Systems invokes it with any relevant parameters from the Nsight Systems run.
The screenshot below shows NCU UI invoked by Nsight Systems. The red circles indicate the parameters pre-populated by Nsight Systems. Users may modify any of these parameters before launching the application and profiling the selected kernel.
CUDA GPU Memory Allocation Graph
When the Collect GPU Memory Usage option is selected from the Collect CUDA trace option set, Nsight Systems will track CUDA GPU memory allocations and deallocations and present a graph of this information in the timeline. This is not the same as the GPU memory graph generated during stutter analysis on the Windows target. See Windows GPU Memory Utilization
Below, in the report on the left, memory is allocated and freed during the collection. In the report on the right, memory is allocated, but not freed during the collection.
Here is another example, where allocations are happening on multiple GPUs
Unified Memory Transfer Trace
For Nsight Systems Workstation Edition, Unified Memory (also called Managed Memory) transfer trace is enabled automatically in Nsight Systems when CUDA trace is selected. It incurs no overhead in programs that do not perform any Unified Memory transfers. Data is displayed in the Managed Memory area of the timeline:
HtoD transfer indicates the CUDA kernel accessed managed memory that was residing on the host, so the kernel execution paused and transferred the data to the device. Heavy traffic here will incur performance penalties in CUDA kernels, so consider using manual cudaMemcpy operations from pinned host memory instead.
PtoP transfer indicates the CUDA kernel accessed managed memory that was residing on a different device, so the kernel execution paused and transferred the data to this device. Heavy traffic here will incur performance penalties, so consider using manual cudaMemcpyPeer operations to transfer from other devices’ memory instead. The row showing these events is for the destination device - the source device is shown in the tooltip for each transfer event.
DtoH transfer indicates the CPU accessed managed memory that was residing on a CUDA device, so the CPU execution paused and transferred the data to system memory. Heavy traffic here will incur performance penalties in CPU code, so consider using manual cudaMemcpy operations from pinned host memory instead.
Some Unified Memory transfers are highlighted with red to indicate potential performance issues:
Transfers with the following migration causes are highlighted:
Coherence
Unified Memory migration occurred to guarantee data coherence. SMs (streaming multiprocessors) stop until the migration completes.
Eviction
Unified Memory migrated to the CPU because it was evicted to make room for another block of memory on the GPU. This happens due to memory overcommitment which is available on Linux with Compute Capability ≥ 6.
Unified Memory CPU Page Faults
The Unified Memory CPU page faults feature in Nsight Systems tracks the page faults that occur when CPU code tries to access a memory page that resides on the device.
Note
Collecting Unified Memory CPU page faults can cause overhead of up to 70% in testing. Please use this functionality only when needed.
Unified Memory GPU Page Faults
The Unified Memory GPU page faults feature in Nsight Systems tracks the page faults that occur when GPU code tries to access a memory page that resides on the host.
Note
Collecting Unified Memory GPU page faults can cause overhead of up to 70% in testing. Please use this functionality only when needed.
CUDA Graph Trace
Nsight Systems is capable of capturing information about CUDA graphs in your
application at either the graph or node granularity. This can be set in the CLI
using the --cuda-graph-trace option, or in the GUI by setting the appropriate
drop down.
When CUDA graph trace is set to graph, the users sees each graph as one item
on the timeline:
When CUDA graph trace is set to node, the users sees each graph as a set of
nodes on the timeline:
Tracing CUDA graphs at the graph level rather than the tracing the underlying nodes results in significantly less overhead. This option is only available ith CUDA driver 515.43 or higher.
CUDA Python Backtrace
Nsight Systems for Arm server (SBSA) platforms and x86 Linux targets, is capable of capturing Python backtrace information when CUDA backtrace is being captured.
To enable CUDA Python backtrace from Nsight Systems:
CLI — Set --python-backtrace=cuda.
GUI — Select the Collect Python backtrace for selected API calls checkbox.
Example screenshot:
CUDA Default Function List for CLI
CUDA Runtime API
cudaBindSurfaceToArray
cudaBindTexture
cudaBindTexture2D
cudaBindTextureToArray
cudaBindTextureToMipmappedArray
cudaConfigureCall
cudaCreateSurfaceObject
cudaCreateTextureObject
cudaD3D10MapResources
cudaD3D10RegisterResource
cudaD3D10UnmapResources
cudaD3D10UnregisterResource
cudaD3D9MapResources
cudaD3D9MapVertexBuffer
cudaD3D9RegisterResource
cudaD3D9RegisterVertexBuffer
cudaD3D9UnmapResources
cudaD3D9UnmapVertexBuffer
cudaD3D9UnregisterResource
cudaD3D9UnregisterVertexBuffer
cudaDestroySurfaceObject
cudaDestroyTextureObject
cudaDeviceReset
cudaDeviceSynchronize
cudaEGLStreamConsumerAcquireFrame
cudaEGLStreamConsumerConnect
cudaEGLStreamConsumerConnectWithFlags
cudaEGLStreamConsumerDisconnect
cudaEGLStreamConsumerReleaseFrame
cudaEGLStreamConsumerReleaseFrame
cudaEGLStreamProducerConnect
cudaEGLStreamProducerDisconnect
cudaEGLStreamProducerReturnFrame
cudaEventCreate
cudaEventCreateFromEGLSync
cudaEventCreateWithFlags
cudaEventDestroy
cudaEventQuery
cudaEventRecord
cudaEventRecord_ptsz
cudaEventSynchronize
cudaFree
cudaFreeArray
cudaFreeHost
cudaFreeMipmappedArray
cudaGLMapBufferObject
cudaGLMapBufferObjectAsync
cudaGLRegisterBufferObject
cudaGLUnmapBufferObject
cudaGLUnmapBufferObjectAsync
cudaGLUnregisterBufferObject
cudaGraphicsD3D10RegisterResource
cudaGraphicsD3D11RegisterResource
cudaGraphicsD3D9RegisterResource
cudaGraphicsEGLRegisterImage
cudaGraphicsGLRegisterBuffer
cudaGraphicsGLRegisterImage
cudaGraphicsMapResources
cudaGraphicsUnmapResources
cudaGraphicsUnregisterResource
cudaGraphicsVDPAURegisterOutputSurface
cudaGraphicsVDPAURegisterVideoSurface
cudaHostAlloc
cudaHostRegister
cudaHostUnregister
cudaLaunch
cudaLaunchCooperativeKernel
cudaLaunchCooperativeKernelMultiDevice
cudaLaunchCooperativeKernel_ptsz
cudaLaunchKernel
cudaLaunchKernel_ptsz
cudaLaunch_ptsz
cudaMalloc
cudaMalloc3D
cudaMalloc3DArray
cudaMallocArray
cudaMallocHost
cudaMallocManaged
cudaMallocMipmappedArray
cudaMallocPitch
cudaMemGetInfo
cudaMemPrefetchAsync
cudaMemPrefetchAsync_ptsz
cudaMemcpy
cudaMemcpy2D
cudaMemcpy2DArrayToArray
cudaMemcpy2DArrayToArray_ptds
cudaMemcpy2DAsync
cudaMemcpy2DAsync_ptsz
cudaMemcpy2DFromArray
cudaMemcpy2DFromArrayAsync
cudaMemcpy2DFromArrayAsync_ptsz
cudaMemcpy2DFromArray_ptds
cudaMemcpy2DToArray
cudaMemcpy2DToArrayAsync
cudaMemcpy2DToArrayAsync_ptsz
cudaMemcpy2DToArray_ptds
cudaMemcpy2D_ptds
cudaMemcpy3D
cudaMemcpy3DAsync
cudaMemcpy3DAsync_ptsz
cudaMemcpy3DPeer
cudaMemcpy3DPeerAsync
cudaMemcpy3DPeerAsync_ptsz
cudaMemcpy3DPeer_ptds
cudaMemcpy3D_ptds
cudaMemcpyArrayToArray
cudaMemcpyArrayToArray_ptds
cudaMemcpyAsync
cudaMemcpyAsync_ptsz
cudaMemcpyFromArray
cudaMemcpyFromArrayAsync
cudaMemcpyFromArrayAsync_ptsz
cudaMemcpyFromArray_ptds
cudaMemcpyFromSymbol
cudaMemcpyFromSymbolAsync
cudaMemcpyFromSymbolAsync_ptsz
cudaMemcpyFromSymbol_ptds
cudaMemcpyPeer
cudaMemcpyPeerAsync
cudaMemcpyToArray
cudaMemcpyToArrayAsync
cudaMemcpyToArrayAsync_ptsz
cudaMemcpyToArray_ptds
cudaMemcpyToSymbol
cudaMemcpyToSymbolAsync
cudaMemcpyToSymbolAsync_ptsz
cudaMemcpyToSymbol_ptds
cudaMemcpy_ptds
cudaMemset
cudaMemset2D
cudaMemset2DAsync
cudaMemset2DAsync_ptsz
cudaMemset2D_ptds
cudaMemset3D
cudaMemset3DAsync
cudaMemset3DAsync_ptsz
cudaMemset3D_ptds
cudaMemsetAsync
cudaMemsetAsync_ptsz
cudaMemset_ptds
cudaPeerRegister
cudaPeerUnregister
cudaStreamAddCallback
cudaStreamAddCallback_ptsz
cudaStreamAttachMemAsync
cudaStreamAttachMemAsync_ptsz
cudaStreamCreate
cudaStreamCreateWithFlags
cudaStreamCreateWithPriority
cudaStreamDestroy
cudaStreamQuery
cudaStreamQuery_ptsz
cudaStreamSynchronize
cudaStreamSynchronize_ptsz
cudaStreamWaitEvent
cudaStreamWaitEvent_ptsz
cudaThreadSynchronize
cudaUnbindTexture
CUDA Primary API
cu64Array3DCreate
cu64ArrayCreate
cu64D3D9MapVertexBuffer
cu64GLMapBufferObject
cu64GLMapBufferObjectAsync
cu64MemAlloc
cu64MemAllocPitch
cu64MemFree
cu64MemGetInfo
cu64MemHostAlloc
cu64Memcpy2D
cu64Memcpy2DAsync
cu64Memcpy2DUnaligned
cu64Memcpy3D
cu64Memcpy3DAsync
cu64MemcpyAtoD
cu64MemcpyDtoA
cu64MemcpyDtoD
cu64MemcpyDtoDAsync
cu64MemcpyDtoH
cu64MemcpyDtoHAsync
cu64MemcpyHtoD
cu64MemcpyHtoDAsync
cu64MemsetD16
cu64MemsetD16Async
cu64MemsetD2D16
cu64MemsetD2D16Async
cu64MemsetD2D32
cu64MemsetD2D32Async
cu64MemsetD2D8
cu64MemsetD2D8Async
cu64MemsetD32
cu64MemsetD32Async
cu64MemsetD8
cu64MemsetD8Async
cuArray3DCreate
cuArray3DCreate_v2
cuArrayCreate
cuArrayCreate_v2
cuArrayDestroy
cuBinaryFree
cuCompilePtx
cuCtxCreate
cuCtxCreate_v2
cuCtxDestroy
cuCtxDestroy_v2
cuCtxSynchronize
cuD3D10CtxCreate
cuD3D10CtxCreateOnDevice
cuD3D10CtxCreate_v2
cuD3D10MapResources
cuD3D10RegisterResource
cuD3D10UnmapResources
cuD3D10UnregisterResource
cuD3D11CtxCreate
cuD3D11CtxCreateOnDevice
cuD3D11CtxCreate_v2
cuD3D9CtxCreate
cuD3D9CtxCreateOnDevice
cuD3D9CtxCreate_v2
cuD3D9MapResources
cuD3D9MapVertexBuffer
cuD3D9MapVertexBuffer_v2
cuD3D9RegisterResource
cuD3D9RegisterVertexBuffer
cuD3D9UnmapResources
cuD3D9UnmapVertexBuffer
cuD3D9UnregisterResource
cuD3D9UnregisterVertexBuffer
cuEGLStreamConsumerAcquireFrame
cuEGLStreamConsumerConnect
cuEGLStreamConsumerConnectWithFlags
cuEGLStreamConsumerDisconnect
cuEGLStreamConsumerReleaseFrame
cuEGLStreamProducerConnect
cuEGLStreamProducerDisconnect
cuEGLStreamProducerPresentFrame
cuEGLStreamProducerReturnFrame
cuEventCreate
cuEventCreateFromEGLSync
cuEventCreateFromNVNSync
cuEventDestroy
cuEventDestroy_v2
cuEventQuery
cuEventRecord
cuEventRecord_ptsz
cuEventSynchronize
cuGLCtxCreate
cuGLCtxCreate_v2
cuGLInit
cuGLMapBufferObject
cuGLMapBufferObjectAsync
cuGLMapBufferObjectAsync_v2
cuGLMapBufferObjectAsync_v2_ptsz
cuGLMapBufferObject_v2
cuGLMapBufferObject_v2_ptds
cuGLRegisterBufferObject
cuGLUnmapBufferObject
cuGLUnmapBufferObjectAsync
cuGLUnregisterBufferObject
cuGraphicsD3D10RegisterResource
cuGraphicsD3D11RegisterResource
cuGraphicsD3D9RegisterResource
cuGraphicsEGLRegisterImage
cuGraphicsGLRegisterBuffer
cuGraphicsGLRegisterImage
cuGraphicsMapResources
cuGraphicsMapResources_ptsz
cuGraphicsUnmapResources
cuGraphicsUnmapResources_ptsz
cuGraphicsUnregisterResource
cuGraphicsVDPAURegisterOutputSurface
cuGraphicsVDPAURegisterVideoSurface
cuInit
cuLaunch
cuLaunchCooperativeKernel
cuLaunchCooperativeKernelMultiDevice
cuLaunchCooperativeKernel_ptsz
cuLaunchGrid
cuLaunchGridAsync
cuLaunchKernel
cuLaunchKernel_ptsz
cuLinkComplete
cuLinkCreate
cuLinkCreate_v2
cuLinkDestroy
cuMemAlloc
cuMemAllocHost
cuMemAllocHost_v2
cuMemAllocManaged
cuMemAllocPitch
cuMemAllocPitch_v2
cuMemAlloc_v2
cuMemFree
cuMemFreeHost
cuMemFree_v2
cuMemGetInfo
cuMemGetInfo_v2
cuMemHostAlloc
cuMemHostAlloc_v2
cuMemHostRegister
cuMemHostRegister_v2
cuMemHostUnregister
cuMemPeerRegister
cuMemPeerUnregister
cuMemPrefetchAsync
cuMemPrefetchAsync_ptsz
cuMemcpy
cuMemcpy2D
cuMemcpy2DAsync
cuMemcpy2DAsync_v2
cuMemcpy2DAsync_v2_ptsz
cuMemcpy2DUnaligned
cuMemcpy2DUnaligned_v2
cuMemcpy2DUnaligned_v2_ptds
cuMemcpy2D_v2
cuMemcpy2D_v2_ptds
cuMemcpy3D
cuMemcpy3DAsync
cuMemcpy3DAsync_v2
cuMemcpy3DAsync_v2_ptsz
cuMemcpy3DPeer
cuMemcpy3DPeerAsync
cuMemcpy3DPeerAsync_ptsz
cuMemcpy3DPeer_ptds
cuMemcpy3D_v2
cuMemcpy3D_v2_ptds
cuMemcpyAsync
cuMemcpyAsync_ptsz
cuMemcpyAtoA
cuMemcpyAtoA_v2
cuMemcpyAtoA_v2_ptds
cuMemcpyAtoD
cuMemcpyAtoD_v2
cuMemcpyAtoD_v2_ptds
cuMemcpyAtoH
cuMemcpyAtoHAsync
cuMemcpyAtoHAsync_v2
cuMemcpyAtoHAsync_v2_ptsz
cuMemcpyAtoH_v2
cuMemcpyAtoH_v2_ptds
cuMemcpyDtoA
cuMemcpyDtoA_v2
cuMemcpyDtoA_v2_ptds
cuMemcpyDtoD
cuMemcpyDtoDAsync
cuMemcpyDtoDAsync_v2
cuMemcpyDtoDAsync_v2_ptsz
cuMemcpyDtoD_v2
cuMemcpyDtoD_v2_ptds
cuMemcpyDtoH
cuMemcpyDtoHAsync
cuMemcpyDtoHAsync_v2
cuMemcpyDtoHAsync_v2_ptsz
cuMemcpyDtoH_v2
cuMemcpyDtoH_v2_ptds
cuMemcpyHtoA
cuMemcpyHtoAAsync
cuMemcpyHtoAAsync_v2
cuMemcpyHtoAAsync_v2_ptsz
cuMemcpyHtoA_v2
cuMemcpyHtoA_v2_ptds
cuMemcpyHtoD
cuMemcpyHtoDAsync
cuMemcpyHtoDAsync_v2
cuMemcpyHtoDAsync_v2_ptsz
cuMemcpyHtoD_v2
cuMemcpyHtoD_v2_ptds
cuMemcpyPeer
cuMemcpyPeerAsync
cuMemcpyPeerAsync_ptsz
cuMemcpyPeer_ptds
cuMemcpy_ptds
cuMemcpy_v2
cuMemsetD16
cuMemsetD16Async
cuMemsetD16Async_ptsz
cuMemsetD16_v2
cuMemsetD16_v2_ptds
cuMemsetD2D16
cuMemsetD2D16Async
cuMemsetD2D16Async_ptsz
cuMemsetD2D16_v2
cuMemsetD2D16_v2_ptds
cuMemsetD2D32
cuMemsetD2D32Async
cuMemsetD2D32Async_ptsz
cuMemsetD2D32_v2
cuMemsetD2D32_v2_ptds
cuMemsetD2D8
cuMemsetD2D8Async
cuMemsetD2D8Async_ptsz
cuMemsetD2D8_v2
cuMemsetD2D8_v2_ptds
cuMemsetD32
cuMemsetD32Async
cuMemsetD32Async_ptsz
cuMemsetD32_v2
cuMemsetD32_v2_ptds
cuMemsetD8
cuMemsetD8Async
cuMemsetD8Async_ptsz
cuMemsetD8_v2
cuMemsetD8_v2_ptds
cuMipmappedArrayCreate
cuMipmappedArrayDestroy
cuModuleLoad
cuModuleLoadData
cuModuleLoadDataEx
cuModuleLoadFatBinary
cuModuleUnload
cuStreamAddCallback
cuStreamAddCallback_ptsz
cuStreamAttachMemAsync
cuStreamAttachMemAsync_ptsz
cuStreamBatchMemOp
cuStreamBatchMemOp_ptsz
cuStreamCreate
cuStreamCreateWithPriority
cuStreamDestroy
cuStreamDestroy_v2
cuStreamSynchronize
cuStreamSynchronize_ptsz
cuStreamWaitEvent
cuStreamWaitEvent_ptsz
cuStreamWaitValue32
cuStreamWaitValue32_ptsz
cuStreamWaitValue64
cuStreamWaitValue64_ptsz
cuStreamWriteValue32
cuStreamWriteValue32_ptsz
cuStreamWriteValue64
cuStreamWriteValue64_ptsz
cuSurfObjectCreate
cuSurfObjectDestroy
cuSurfRefCreate
cuSurfRefDestroy
cuTexObjectCreate
cuTexObjectDestroy
cuTexRefCreate
cuTexRefDestroy
cuVDPAUCtxCreate
cuVDPAUCtxCreate_v2
cuDNN Function List for X86 CLI
cuDNN API functions
cudnnActivationBackward
cudnnActivationBackward_v3
cudnnActivationBackward_v4
cudnnActivationForward
cudnnActivationForward_v3
cudnnActivationForward_v4
cudnnAddTensor
cudnnBatchNormalizationBackward
cudnnBatchNormalizationBackwardEx
cudnnBatchNormalizationForwardInference
cudnnBatchNormalizationForwardTraining
cudnnBatchNormalizationForwardTrainingEx
cudnnCTCLoss
cudnnConvolutionBackwardBias
cudnnConvolutionBackwardData
cudnnConvolutionBackwardFilter
cudnnConvolutionBiasActivationForward
cudnnConvolutionForward
cudnnCreate
cudnnCreateAlgorithmPerformance
cudnnDestroy
cudnnDestroyAlgorithmPerformance
cudnnDestroyPersistentRNNPlan
cudnnDivisiveNormalizationBackward
cudnnDivisiveNormalizationForward
cudnnDropoutBackward
cudnnDropoutForward
cudnnDropoutGetReserveSpaceSize
cudnnDropoutGetStatesSize
cudnnFindConvolutionBackwardDataAlgorithm
cudnnFindConvolutionBackwardDataAlgorithmEx
cudnnFindConvolutionBackwardFilterAlgorithm
cudnnFindConvolutionBackwardFilterAlgorithmEx
cudnnFindConvolutionForwardAlgorithm
cudnnFindConvolutionForwardAlgorithmEx
cudnnFindRNNBackwardDataAlgorithmEx
cudnnFindRNNBackwardWeightsAlgorithmEx
cudnnFindRNNForwardInferenceAlgorithmEx
cudnnFindRNNForwardTrainingAlgorithmEx
cudnnFusedOpsExecute
cudnnIm2Col
cudnnLRNCrossChannelBackward
cudnnLRNCrossChannelForward
cudnnMakeFusedOpsPlan
cudnnMultiHeadAttnBackwardData
cudnnMultiHeadAttnBackwardWeights
cudnnMultiHeadAttnForward
cudnnOpTensor
cudnnPoolingBackward
cudnnPoolingForward
cudnnRNNBackwardData
cudnnRNNBackwardDataEx
cudnnRNNBackwardWeights
cudnnRNNBackwardWeightsEx
cudnnRNNForwardInference
cudnnRNNForwardInferenceEx
cudnnRNNForwardTraining
cudnnRNNForwardTrainingEx
cudnnReduceTensor
cudnnReorderFilterAndBias
cudnnRestoreAlgorithm
cudnnRestoreDropoutDescriptor
cudnnSaveAlgorithm
cudnnScaleTensor
cudnnSoftmaxBackward
cudnnSoftmaxForward
cudnnSpatialTfGridGeneratorBackward
cudnnSpatialTfGridGeneratorForward
cudnnSpatialTfSamplerBackward
cudnnSpatialTfSamplerForward
cudnnTransformFilter
cudnnTransformTensor
cudnnTransformTensorEx
OpenACC Trace
Nsight Systems for Linux x86_64 is capable of capturing information about OpenACC execution in the profiled process.
OpenACC versions 2.0, 2.5, and 2.6 are supported when using PGI runtime version 15.7 or later. In order to differentiate constructs (see tooltip below), a PGI runtime of 16.0 or later is required. Note that Nsight Systems does not support the GCC implementation of OpenACC at this time.
Under the CPU rows in the timeline tree, each thread that uses OpenACC will show OpenACC trace information. You can click on a OpenACC API call to see correlation with the underlying CUDA API calls (highlighted in teal):
If the OpenACC API results in GPU work, that will also be highlighted:
Hovering over a particular OpenACC construct will bring up a tooltip with details about that construct:
To capture OpenACC information from the Nsight Systems GUI, select the Collect OpenACC trace checkbox under Collect CUDA trace configurations. Note that turning on OpenACC tracing will also turn on CUDA tracing.
Please note that if your application crashes before all collected OpenACC trace data has been copied out, some or all data might be lost and not present in the report.
OpenGL Trace
OpenGL and OpenGL ES APIs can be traced to assist in the analysis of CPU and GPU interactions.
A few usage examples are:
Visualize how long
eglSwapBuffers(or similar) is taking.API trace can easily show correlations between thread state and graphics driver’s behavior, uncovering where the CPU may be waiting on the GPU.
Spot bubbles of opportunity on the GPU, where more GPU workload could be created.
Use
KHR_debugextension to trace GL events on both the CPU and GPU.
OpenGL trace feature in Nsight Systems consists of two different activities which will be shown in the CPU rows for those threads
CPU trace: interception of API calls that an application does to APIs (such as OpenGL, OpenGL ES, EGL, GLX, WGL, etc.).
GPU trace (or workload trace): trace of GPU workload (activity) triggered by use of OpenGL or OpenGL ES. Since draw calls are executed back-to-back, the GPU workload trace ranges include many OpenGL draw calls and operations in order to optimize performance overhead, rather than tracing each individual operation.
To collect GPU trace, the glQueryCounter() function is used to measure how much time batches of GPU workload take to complete.
Ranges defined by the KHR_debug calls are represented similarly to OpenGL API and OpenGL GPU workload trace. GPU ranges in this case represent incremental draw cost. They cannot fully account for GPUs that can execute multiple draw calls in parallel. In this case, Nsight Systems will not show overlapping GPU ranges.
OpenGL Trace Using Command Line
For general information on using the target CLI, see CLI Profiling on Linux. For the CLI, the functions that are traced are set to the following list:
glWaitSync
glReadPixels
glReadnPixelsKHR
glReadnPixelsEXT
glReadnPixelsARB
glReadnPixels
glFlush
glFinishFenceNV
glFinish
glClientWaitSync
glClearTexSubImage
glClearTexImage
glClearStencil
glClearNamedFramebufferuiv
glClearNamedFramebufferiv
glClearNamedFramebufferfv
glClearNamedFramebufferfi
glClearNamedBufferSubDataEXT
glClearNamedBufferSubData
glClearNamedBufferDataEXT
glClearNamedBufferData
glClearIndex
glClearDepthx
glClearDepthf
glClearDepthdNV
glClearDepth
glClearColorx
glClearColorIuiEXT
glClearColorIiEXT
glClearColor
glClearBufferuiv
glClearBufferSubData
glClearBufferiv
glClearBufferfv
glClearBufferfi
glClearBufferData
glClearAccum
glClear
glDispatchComputeIndirect
glDispatchComputeGroupSizeARB
glDispatchCompute
glComputeStreamNV
glNamedFramebufferDrawBuffers
glNamedFramebufferDrawBuffer
glMultiDrawElementsIndirectEXT
glMultiDrawElementsIndirectCountARB
glMultiDrawElementsIndirectBindlessNV
glMultiDrawElementsIndirectBindlessCountNV
glMultiDrawElementsIndirectAMD
glMultiDrawElementsIndirect
glMultiDrawElementsEXT
glMultiDrawElementsBaseVertex
glMultiDrawElements
glMultiDrawArraysIndirectEXT
glMultiDrawArraysIndirectCountARB
glMultiDrawArraysIndirectBindlessNV
glMultiDrawArraysIndirectBindlessCountNV
glMultiDrawArraysIndirectAMD
glMultiDrawArraysIndirect
glMultiDrawArraysEXT
glMultiDrawArrays
glListDrawCommandsStatesClientNV
glFramebufferDrawBuffersEXT
glFramebufferDrawBufferEXT
glDrawTransformFeedbackStreamInstanced
glDrawTransformFeedbackStream
glDrawTransformFeedbackNV
glDrawTransformFeedbackInstancedEXT
glDrawTransformFeedbackInstanced
glDrawTransformFeedbackEXT
glDrawTransformFeedback
glDrawTexxvOES
glDrawTexxOES
glDrawTextureNV
glDrawTexsvOES
glDrawTexsOES
glDrawTexivOES
glDrawTexiOES
glDrawTexfvOES
glDrawTexfOES
glDrawRangeElementsEXT
glDrawRangeElementsBaseVertexOES
glDrawRangeElementsBaseVertexEXT
glDrawRangeElementsBaseVertex
glDrawRangeElements
glDrawPixels
glDrawElementsInstancedNV
glDrawElementsInstancedEXT
glDrawElementsInstancedBaseVertexOES
glDrawElementsInstancedBaseVertexEXT
glDrawElementsInstancedBaseVertexBaseInstanceEXT
glDrawElementsInstancedBaseVertexBaseInstance
glDrawElementsInstancedBaseVertex
glDrawElementsInstancedBaseInstanceEXT
glDrawElementsInstancedBaseInstance
glDrawElementsInstancedARB
glDrawElementsInstanced
glDrawElementsIndirect
glDrawElementsBaseVertexOES
glDrawElementsBaseVertexEXT
glDrawElementsBaseVertex
glDrawElements
glDrawCommandsStatesNV
glDrawCommandsStatesAddressNV
glDrawCommandsNV
glDrawCommandsAddressNV
glDrawBuffersNV
glDrawBuffersATI
glDrawBuffersARB
glDrawBuffers
glDrawBuffer
glDrawArraysInstancedNV
glDrawArraysInstancedEXT
glDrawArraysInstancedBaseInstanceEXT
glDrawArraysInstancedBaseInstance
glDrawArraysInstancedARB
glDrawArraysInstanced
glDrawArraysIndirect
glDrawArraysEXT
glDrawArrays
eglSwapBuffersWithDamageKHR
eglSwapBuffers
glXSwapBuffers
glXQueryDrawable
glXGetCurrentReadDrawable
glXGetCurrentDrawable
glGetQueryObjectuivEXT
glGetQueryObjectuivARB
glGetQueryObjectuiv
glGetQueryObjectivARB
glGetQueryObjectiv
OpenXR API Trace
OpenXR is a royalty-free, open standard that provides high-performance access to Augmented Reality (AR) and Virtual Reality (VR)—collectively known as XR—platforms and devices. Information about OpenXR can be found at the OpenXR Overview.
Nsight Systems can capture information about OpenXR usage by the profiled process. This includes capturing the execution time of OpenXR API functions, debug labels, and frame durations. OpenXR profiling is supported on Windows operating systems.
Custom ETW Trace
Use the custom ETW trace feature to enable and collect any manifest-based ETW log. The collected events are displayed on the timeline on dedicated rows for each event type.
Custom ETW is available on Windows target machines.
To retain the .etl trace files captured, so that they can be viewed in other tools (e.g. GPUView), change the “Save ETW log files in project folder” option under “Profile Behavior” in Nsight Systems’s global Options dialog. The .etl files will appear in the same folder as the .nsys-rep file, accessible by right-clicking the report in the Project Explorer and choosing “Show in Folder…”. Data collected from each ETW provider will appear in its own .etl file, and an additional .etl file named “Report XX-Merged-*.etl”, containing the events from all captured sources, will be created as well.
GPU Hardware Profiling
GPU Context Switch
Nsight Systems provides the ability to trace GPU context switches. Note that this requires driver r435.17 or later and root permission.
To enable trace, run from the CLI using the --gpuctxsw option
Specifically, the behavior is as follows:
When collecting GPU context switch data as root, you will get records about contexts from all processes. The records have valid context IDs and process IDs, and have full-precision timestamps.
When collecting GPU context switch data as a normal user, you will still get records about contexts from all processes. For processes running as your user, the records have valid context ID and process IDs, and full-precision timestamps. For processes running as a different user, the records have context ID = 0 and process ID = 0, and reduced-precision timestamps (which are still guaranteed to be in the correct order).
When collecting GPU context switch data in a virtual machine using vGPU, the above rules apply to records relating to your VM. No records are collected for contexts running on other VMs, so the timeline may show gaps when the vGPU is switched to another VM’s context(s). We do not currently support collecting GPU context switch data on a host system where vGPUs are in use by VMs.
GPU Metrics
Overview
GPU Metrics feature is intended to identify performance limiters in applications using GPU for computations and graphics. It uses periodic sampling to gather performance metrics and detailed timing statistics associated with different GPU hardware units taking advantage of specialized hardware to capture this data in a single pass with minimal overhead.
Note: GPU Metrics will give you precise device level information, but it does not know which process or context is involved. GPU context switch trace provides less precise information, but will give you process and context information.
These metrics provide an overview of GPU efficiency over time within compute, graphics, and input/output (IO) activities such as:
IO throughputs: PCIe, NVLink, and GPU memory bandwidth
SM utilization: SMs activity, tensor core activity, instructions issued, warp occupancy, and unassigned warp slots
It is designed to help users answer the common questions:
Is my GPU idle?
Is my GPU full? Enough kernel grids size and streams? Are my SMs and warp slots full?
Am I using TensorCores?
Is my instruction rate high?
Am I possibly blocked on IO, or number of warps, etc
Nsight Systems GPU Metrics is only available for Linux targets on x86-64 and aarch64, and for Windows targets. It requires NVIDIA Turing architecture or newer.
Minimum required driver versions:
NVIDIA Turing architecture TU10x, TU11x - r440
NVIDIA Ampere architecture GA100 - r450
NVIDIA Ampere architecture GA100 MIG - r470 TRD1
NVIDIA Ampere architecture GA10x - r455
Note
Permissions: Elevated permissions are required. On Linux use sudo to elevate privileges. On Windows the user must run from an admin command prompt or accept the UAC escalation dialog. See Permissions Issues and Performance Counters for more information.
Note
Tensor Core: If you run nsys profile --gpu-metrics-devices all, the
Tensor Core utilization can be found in the GUI under the
SM instructions/Tensor Active row.
Please note that it is not practical to expect a CUDA kernel to reach 100% Tensor Core utilization since there are other overheads. In general, the more computation-intensive an operation is, the higher Tensor Core utilization rate the CUDA kernel can achieve.
Launching GPU Metrics from the CLI
GPU Metrics feature is controlled with 3 CLI switches:
--gpu-metrics-devices=[all, none, <index>]selects GPUs to sample (default is none)--gpu-metrics-set=[<alias>, file:<file name>]selects metric set to use (default is the 1st suitable from the list)--gpu-metrics-frequency=[10..200000]selects sampling frequency in Hz (default is 10000)
To profile with default options and sample GPU Metrics on GPU 1:
# Must have elevated permissions (see https://developer.nvidia.com/ERR_NVGPUCTRPERM) or be root (Linux) or Administrator (Windows)
$ nsys profile --gpu-metrics-devices=1 ./my-app
To list available GPUs, use:
$ nsys profile --gpu-metrics-devices=help
Possible --gpu-metrics-devices values are:
0: Quadro GV100 PCI[0000:17:00.0] (not supported)
1: GeForce RTX 2070 SUPER PCI[0000:65:00.0]
all: Select all supported GPUs
none: Disable GPU Metrics [Default]
By default, the first metric set which supports all selected GPUs is used. But you can manually select another metric set from the list. To see available metric sets, use:
$ nsys profile --gpu-metrics-set=help
Possible --gpu-metrics-set values are:
tu10x : General Metrics for NVIDIA TU10x (any frequency)
tu11x : General Metrics for NVIDIA TU11x (any frequency)
ga100 : General Metrics for NVIDIA GA100 (any frequency)
ga10x : General Metrics for NVIDIA GA10x (any frequency)
ga10b : General Metrics for NVIDIA GA10B (any frequency)
gh100 : General Metrics for NVIDIA GH100 (any frequency)
ad10x : General Metrics for NVIDIA AD10x (any frequency)
tu10x-gfxt : Graphics Throughput Metrics for NVIDIA TU10x (frequency >= 10kHz)
ga10x-gfxt : Graphics Throughput Metrics for NVIDIA GA10x (frequency >= 10kHz)
ad10x-gfxt : Graphics Throughput Metrics for NVIDIA AD10x (frequency >= 10kHz)
ga10x-gfxact : Graphics Async Compute Triage Metrics for NVIDIA GA10x (frequency >= 10kHz)
file:<file name> : use metric set from a given file
By default, sampling frequency is set to 10 kHz. But you can manually set it from 10 Hz to 200 kHz using
--gpu-metrics-frequency=<value>
Launching GPU Metrics from the GUI
For commands to launch GPU Metrics from the CLI with examples, see Profiling from the CLI
When launching analysis in Nsight Systems, select Collect GPU Metrics.
Select the GPUs dropdown to pick which GPUs you wish to sample.
Select the Metric set: dropdown to choose which available metric set you would like to sample.
Note that metric sets for GPUs that are not being sampled will be greyed out.
Sampling frequency
Sampling frequency can be selected from the range of 10 Hz - 200 kHz. The default value is 10 kHz.
The maximum sampling frequency without buffer overflow events depends on GPU (SM count), GPU load intensity, and overall system load. The bigger the chip and the higher the load, the lower the maximum frequency. If you need higher frequency, you can increase it until you get “Buffer overflow” message in the Diagnostics Summary report page.
Each metric set has a recommended sampling frequency range in its description.
These ranges are approximate. If you observe Inconsistent Data or
Missing Data ranges on timeline, please try closer to the recommended
frequency.
Available metrics
GPC Clock Frequency -
gpc__cycles_elapsed.avg.per_secondThe average GPC clock frequency in hertz. In public documentation the GPC clock may be called the “Application” clock, “Graphic” clock, “Base” clock, or “Boost” clock.
Note: The collection mechanism for GPC can result in a small fluctuation between samples.
SYS Clock Frequency -
sys__cycles_elapsed.avg.per_secondThe average SYS clock frequency in hertz. The GPU front end (command processor), copy engines, and the performance monitor run at the SYS clock. On Turing and NVIDIA GA100 GPUs the sampling frequency is based upon a period of SYS clocks (not time) so samples per second will vary with SYS clock. On NVIDIA GA10x GPUs the sampling frequency is based upon a fixed frequency clock. The maximum frequency scales linearly with the SYS clock.
GR Active -
gr__cycles_active.sum.pct_of_peak_sustained_elapsedThe percentage of cycles the graphics/compute engine is active. The graphics/compute engine is active if there is any work in the graphics pipe or if the compute pipe is processing work.
GA100 MIG - MIG is not yet supported. This counter will report the activity of the primary GR engine.
Sync Compute In Flight -
gr__dispatch_cycles_active_queue_sync.avg.pct_of_peak_sustained_elapsedThe percentage of cycles with synchronous compute in flight.
CUDA: CUDA will only report synchronous queue in the case of MPS configured with 64 sub-context. Synchronous refers to work submitted in VEID=0.
Graphics: This will be true if any compute work submitted from the direct queue is in flight.
Async Compute in Flight -
gr__dispatch_cycles_active_queue_async.avg.pct_of_peak_sustained_elapsedThe percentage of cycles with asynchronous compute in flight.
CUDA: CUDA will only report all compute work as asynchronous. The one exception is if MPS is configured and all 64 sub-context are in use. 1 sub-context (VEID=0) will report as synchronous.
Graphics: This will be true if any compute work submitted from a compute queue is in flight.
Draw Started -
fe__draw_count.avg.pct_of_peak_sustained_elapsedThe ratio of draw calls issued to the graphics pipe to the maximum sustained rate of the graphics pipe.
Note:The percentage will always be very low as the front end can issue draw calls significantly faster than the pipe can execute the draw call. The rendering of this row will be changed to help indicate when draw calls are being issued.
Dispatch Started -
gr__dispatch_count.avg.pct_of_peak_sustained_elapsedThe ratio of compute grid launches (dispatches) to the compute pipe to the maximum sustained rate of the compute pipe.
Note: The percentage will always be very low as the front end can issue grid launches significantly faster than the pipe can execute the draw call. The rendering of this row will be changed to help indicate when grid launches are being issued.
Vertex/Tess/Geometry Warps in Flight -
tpc__warps_active_shader_vtg_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of active vertex, geometry, tessellation, and meshlet shader warps resident on the SMs to the maximum number of warps per SM as a percentage.
Pixel Warps in Flight -
tpc__warps_active_shader_ps_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of active pixel/fragment shader warps resident on the SMs to the maximum number of warps per SM as a percentage.
Compute Warps in Flight -
tpc__warps_active_shader_cs_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of active compute shader warps resident on the SMs to the maximum number of warps per SM as a percentage.
Active SM Unused Warp Slots -
tpc__warps_inactive_sm_active_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of inactive warp slots on the SMs to the maximum number of warps per SM as a percentage. This is an indication of how many more warps may fit on the SMs if occupancy is not limited by a resource such as max warps of a shader type, shared memory, registers per thread, or thread blocks per SM.
Idle SM Unused Warp Slots -
tpc__warps_inactive_sm_idle_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of inactive warps slots due to idle SMs to the the maximum number of warps per SM as a percentage.
This is an indicator that the current workload on the SM is not sufficient to put work on all SMs. This can be due to:
CPU starving the GPU
current work is too small to saturate the GPU
current work is trailing off but blocking next work
SMs Active -
sm__cycles_active.avg.pct_of_peak_sustained_elapsedThe ratio of cycles SMs had at least 1 warp in flight (allocated on SM) to the number of cycles as a percentage. A value of 0 indicates all SMs were idle (no warps in flight). A value of 50% can indicate some gradient between all SMs active 50% of the sample period or 50% of SMs active 100% of the sample period.
SM Issue -
sm__inst_executed_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of cycles that SM sub-partitions (warp schedulers) issued an instruction to the number of cycles in the sample period as a percentage.
Tensor Active -
sm__pipe_tensor_cycles_active_realtime.avg.pct_of_peak_sustained_elapsedThe ratio of cycles the SM tensor pipes were active issuing tensor instructions to the number of cycles in the sample period as a percentage.
TU102/4/6: This metric is not available on TU10x for periodic sampling. Please see Tensor Active/FP16 Active.
Tensor Active / FP16 Active -
sm__pipe_shared_cycles_active_realtime.avg.pct_of_peak_sustained_elapsedTU102/4/6 only
The ratio of cycles the SM tensor pipes or FP16x2 pipes were active issuing tensor instructions to the number of cycles in the sample period as a percentage.
DRAM Read Bandwidth -
dramc__read_throughput.avg.pct_of_peak_sustained_elapsed,dram__read_throughput.avg.pct_of_peak_sustained_elapsedVRAM Read Bandwidth -
FBPA.TriageA.dramc__read_throughput.avg.pct_of_peak_sustained_elapsed,FBSP.TriageSCG.dramc__read_throughput.avg.pct_of_peak_sustained_elapsed,FBSP.TriageAC.dramc__read_throughput.avg.pct_of_peak_sustained_elapsedThe ratio of cycles the DRAM interface was active reading data to the elapsed cycles in the same period as a percentage.
DRAM Write Bandwidth -
dramc__write_throughput.avg.pct_of_peak_sustained_elapsed,dram__write_throughput.avg.pct_of_peak_sustained_elapsedVRAM Write Bandwidth -
FBPA.TriageA.dramc__write_throughput.avg.pct_of_peak_sustained_elapsed,FBSP.TriageSCG.dramc__write_throughput.avg.pct_of_peak_sustained_elapsed,FBSP.TriageAC.dramc__write_throughput.avg.pct_of_peak_sustained_elapsedThe ratio of cycles the DRAM interface was active writing data to the elapsed cycles in the same period as a percentage.
NVLink bytes received -
nvlrx__bytes.avg.pct_of_peak_sustained_elapsedThe ratio of bytes received on the NVLink interface to the maximum number of bytes receivable in the sample period as a percentage. This value includes protocol overhead.
NVLink bytes transmitted -
nvltx__bytes.avg.pct_of_peak_sustained_elapsedThe ratio of bytes transmitted on the NVLink interface to the maximum number of bytes transmittable in the sample period as a percentage. This value includes protocol overhead.
PCIe Read Throughput -
pcie__read_bytes.avg.pct_of_peak_sustained_elapsedThe ratio of bytes received on the PCIe interface to the maximum number of bytes receivable in the sample period as a percentage. The theoretical value is calculated based upon the PCIe generation and number of lanes. This value includes protocol overhead.
PCIe Write Throughput -
pcie__write_bytes.avg.pct_of_peak_sustained_elapsedThe ratio of bytes transmitted on the PCIe interface to the maximum number of bytes receivable in the sample period as a percentage. The theoretical value is calculated based upon the PCIe generation and number of lanes. This value includes protocol overhead.
PCIe Read Requests to BAR1 -
pcie__rx_requests_aperture_bar1_op_read.sumPCIe Write Requests to BAR1 -
pcie__rx_requests_aperture_bar1_op_write.sumBAR1 is a PCI Express (PCIe) interface used to allow the CPU or other devices to directly access GPU memory. The GPU normally transfers memory with its copy engines, which would not show up as BAR1 activity. The GPU drivers on the CPU do a small amount of BAR1 accesses, but heavier traffic is typically coming from other technologies.
On Linux, technologies like GPU Direct, GPU Direct RDMA, and GPU Direct Storage transfer data across PCIe BAR1. In the case of GPU Direct RDMA, that would be an Ethernet or InfiniBand adapter directly writing to GPU memory.
On Windows, Direct3D12 resources can also be made accessible directly to the CPU via NVAPI functions to support small writes or reads from GPU buffers, in this case too many BAR1 accesses can indicate a performance issue, like it has been demonstrated in the Optimizing DX12 Resource Uploads to the GPU Using CPU-Visible VRAM technical blog post.
Exporting and Querying Data
It is possible to access metric values for automated processing using the Nsight Systems CLI export capabilities.
An example that extracts values of “SMs Active”:
$ nsys export -t sqlite report.nsys-rep
$ sqlite3 report.sqlite "SELECT rawTimestamp, CAST(JSON_EXTRACT(data, '$.\"SMs Active\"') as INTEGER) as value FROM GENERIC_EVENTS WHERE value != 0 LIMIT 10"
309277039|80
309301295|99
309325583|99
309349776|99
309373872|60
309397872|19
309421840|100
309446000|100
309470096|100
309494161|99
An overview of data stored in each event (JSON):
$ sqlite3 report.sqlite "SELECT data FROM GENERIC_EVENTS LIMIT 1"
{
"Unallocated Warps in Active SM": "0",
"Compute Warps In Flight": "52",
"Pixel Warps In Flight": "0",
"Vertex\/Tess\/Geometry Warps In Flight": "0",
"Total SM Occupancy": "52",
"GR Active (GE\/CE)": "100",
"Sync Compute In Flight": "0",
"Async Compute In Flight": "98",
"NVLink bytes received": "0",
"NVLink bytes transmitted": "0",
"PCIe Rx Throughput": "0",
"PCIe Tx Throughput": "1",
"DRAM Read Throughput": "0",
"DRAM Write Throughput": "0",
"Tensor Active \/ FP16 Active": "0",
"SM Issue": "10",
"SMs Active": "52"
}
Values are integer percentages (0..100)
Limitations
If metric sets with NVLink are used but the links are not active, they may appear as fully utilized.
Only one tool that subscribes to these counters can be used at a time, therefore, Nsight Systems GPU Metrics feature cannot be used at the same time as the following tools:
Nsight Graphics
Nsight Compute
DCGM (Data Center GPU Manager)
Use the following command:
dcgmi profile --pausedcgmi profile --resume
Or API:
dcgmProfPausedcgmProfResume
Non-NVIDIA products which use:
CUPTI sampling used directly in the application. CUPTI trace is okay (although it will block Nsight Systems CUDA trace)
DCGM library
Nsight Systems limits the amount of memory that can be used to store GPU Metrics samples. Analysis with higher sampling rates or on GPUs with more SMs has a risk of exceeding this limit. This will lead to gaps on timeline filled with
Missing Dataranges. Future releases will reduce the frequency of this happening.
NVML power and temperature metrics (preview)
Nsight Systems can now periodically sample power and temperature metrics from GPUs and plot them on
the timeline in the GUI.
These metrics are provided by the NVML API calls nvmlDeviceGetPowerUsage and
nvmlDeviceGetTemperature respectively. The power metrics are provided in milliwatts (mW) and
the temperature in degrees Celcius (C).
To enable the power and temperature sampling add the following option to the nsys
profile or start commands:
--enable nvml_metrics[,arg1[=value1],arg2[=value2], ...]
There are no spaces following nvml_metrics plugin name. It is followed by a
comma separated list of arguments or argument=value pairs. Arguments with spaces
should be enclosed in double quotes.
Supported arguments are:
Short name |
Long name |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
|
integer |
100000 |
Sampling interval in microseconds |
|
|
Print help message |
Usage Examples
nsys profile --enable nvml_metrics ...Sample power and temperature on all available GPUs every 100ms.
nsys profile --enable nvml_metrics,-i10000Sample power and temperature on all available GPUs every 10ms.
For general information on Nsight Systems plugins please refer to Nsight Systems Plugins (Preview) system.
SoC Metrics
Overview
SoC Metrics feature is intended to identify performance limiters in applications running on NVIDIA SoCs and is similar to GPU Metrics.
Nsight Systems SoC Metrics is only available for Linux and QNX targets on aarch64. It requires NVIDIA Orin architecture or newer.
Available metrics
- CPU Read Throughput
mcc__dram_throughput_srcnode_cpu_op_read.avg.pct_of_peak_sustained_elapsed- CPU Write Throughput
mcc__dram_throughput_srcnode_cpu_op_write.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the SoC memory controllers were actively processing read/write operations from the CPU to the number of cycles in the same sample period as a percentage.
- GPU Read Throughput
mcc__dram_throughput_srcnode_gpu_op_read.avg.pct_of_peak_sustained_elapsed- GPU Write Throughput
mcc__dram_throughput_srcnode_gpu_op_write.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the SoC memory controllers were actively processing read/write operations from the GPU to the number of cycles in the same sample period as a percentage.
- DBB Read Throughput
mcc__dram_throughput_srcnode_dbb_op_read.avg.pct_of_peak_sustained_elapsed- DBB Write Throughput
mcc__dram_throughput_srcnode_dbb_op_write.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the SoC memory controllers were actively processing read/write operations from not-CPU/not-GPU to the number of cycles in the same sample period as a percentage.
- DRAM Read Throughput
mcc__dram_throughput_op_read.avg.pct_of_peak_sustained_elapsed- DRAM Write Throughput
mcc__dram_throughput_op_write.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the SoC memory controllers were actively processing read/write operations to the number of cycles in the same sample period as a percentage.
- DLA0/DLA1 Active
nvdla__cycles_active.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the DLA (Deep Learning Accelerator) was actively processing a command to the number of cycles in the same sample period as a percentage.
- DLA0/DLA1 Read Throughput
nvdla__dbb2nvdla_read_throughput.avg.pct_of_peak_sustained_elapsed- DLA0/DLA1 Write Throughput
nvdla__nvdla2dbb_write_throughput.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the DLA (Deep Learning Accelerator) was actively processing read/write operations to the number of cycles in the same sample period as a percentage.
- NVENC Active
nvenc__cycles_active.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the NVENC unit was actively processing a command to the number of cycles in the same sample period as a percentage.
- NVENC Read Throughput
nvenc__memif2nvenc_read_throughput.avg.pct_of_peak_sustained_elapsed- NVENC Write Throughput
nvenc__nvenc2memif_write_throughput.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the NVENC unit was actively processing read/write operations to the number of cycles in the same sample period as a percentage.
- PVA VPU Active
pvavpu__vpu_cycles_active.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the PVA (Programmable Vision Accelerator) VPU (Vector Processing Unit) was actively processing a command to the number of cycles in the same sample period as a percentage.
- PVA DMA Read Throughput
pva__dbb2pvadma_read_throughput.avg.pct_of_peak_sustained_elapsed- PVA DMA Write Throughput
pva__pvadma2dbb_write_throughput.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the PVA (Programmable Vision Accelerator) VPU (Vector Processing Unit) was actively processing read/write operations to the number of cycles in the same sample period as a percentage.
Note
To enable PVA trace on DRIVE 6.0.8.0, run these two commands before mounting any additional partitions:
echo 1 >/dev/nvpvadebugfs/pva0/tracingecho 2 >/dev/nvpvadebugfs/pva0/trace_level- OFA Active
ofa_cycles_active.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the OFA (Optical Flow Accelerator) was actively processing a command to the number of cycles in the same sample period as a percentage.
- OFA Read Throughput
ofa__memif2ofa_read_throughput.avg.pct_of_peak_sustained_elapsed- OFA Write Throughput
ofa__ofa2memif_write_throughput.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the OFA (Optical Flow Accelerator) was actively processing read/write operations to the number of cycles in the same sample period as a percentage.
- VIC Active
vic_cycles_active.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the VIC (Video Image Compositor) was actively processing a command to the number of cycles in the same sample period as a percentage.
- VIC Read Throughput
vic__dbb2vic_read_throughput.avg.pct_of_peak_sustained_elapsed- VIC Write Throughput
vic__vic2dbb_write_throughput.avg.pct_of_peak_sustained_elapsed
The ratio of cycles the VIC (Video Image Compositor) was actively processing read/write operations to the number of cycles in the same sample period as a percentage.
Launching SoC Metrics from the CLI
SoC Metrics feature is controlled with 3 CLI switches:
--soc-metrics=[true, false]enables SoC Metrics sampling (default is false)--soc-metrics-set=[<alias>, file:<file name>]selects metric set to use (default is the 1st suitable from the list)--soc-metrics-frequency=[100..200000]selects sampling frequency in Hz (default is 10000)
To profile with default options:
# Must have elevated permissions (see https://developer.nvidia.com/ERR_NVGPUCTRPERM) or be root
$ nsys profile --soc-metrics=true ./my-app
Launching SoC Metrics from the GUI
When launching analysis in Nsight Systems, select Collect SoC Metrics.
The settings are similar to GPU Metrics.
For commands to launch SoC Metrics from the CLI with examples, see the CLI documentation.
CPU Profiling on Linux
Nsight Systems on Linux targets, utilizes the Linux OS’ perf subsystem to sample CPU Instruction Pointers (IPs) and backtraces, trace CPU context switches, and sample CPU and OS event counts. The Linux perf tool utilizes the same perf subsystem.
Nsight Systems Embedded Platforms Edition on Linux kernel prior to v5.15 uses a custom kernel module to
collect the same data. The Nsight Systems CLI command
nsys status --environment indicates when the kernel module is used instead
of the Linux OS’ perf subsystem.
Features
CPU Instruction Pointer / Backtrace Sampling
Nsight Systems can sample CPU Instruction Pointers / backtraces periodically. The collection of a sample is triggered by a hardware event overflow - e.g. a sample is collected after every 1 million CPU reference cycles on a per thread basis. In the GUI, samples are shown on the individual thread timelines, in the Event Viewer, and in the Top Down, Bottom Up, or Flat views which provide histogram-like summaries of the data. IP / backtrace collections can be configured in process-tree or system-wide mode. In process-tree mode, Nsight Systems will sample the process, and any of its descendants, launched by the tool. In system-wide mode, Nsight Systems will sample all processes running on the system, including any processes launched by the tool.
CPU Context Switch Tracing
Nsight Systems can trace every time the OS schedules a thread on a logical CPU and every time the OS thread gets unscheduled from a logical CPU. The data is used to show CPU utilization and OS thread utilization within the Nsight Systems GUI. Context switch collections can be configured in process-tree or system-wide mode. In process-tree mode, Nsight Systems will trace the process, and any of its descendants, launched by Nsight Systems. In system-wide mode, Nsight Systems will trace all processes running on the system, including any processes launched by the Nsight Systems.
CPU Event Sampling
Nsight Systems can periodically sample CPU hardware event counts and OS event counts and show the event’s rate over time in the Nsight Systems GUI. Event sample collections can be configured in system-wide mode only. In system-wide mode, Nsight Systems will sample event counts of all CPUs and the OS event counts running on the system. Event counts are not directly associated with processes or threads.
CPU Core Metrics
Nsight Systems can access and make available information about CPU core metrics. This functionality is available only on Linux and only for the NVIDIA Grace CPU. The
--cpu-core-metrics=helpcommand will list 39 different metrics, Those metrics are described in the Grace Performance Tuning Guide. Then selected option IDs can be fed into the--cpu-core-metricsswitch.
System Requirements
Paranoid Level
The system’s paranoid level must be 2 or lower.
Paranoid Level |
CPU IP/backtrace Sampling process-tree mode |
CPU IP/backtrace Sampling system-wide mode |
CPU Context Switch Tracing process-tree mode |
CPU Context Switch Tracing system-wide mode |
Event Sampling system-wide mode |
|---|---|---|---|---|---|
3 or greater |
not available |
not available |
not available |
not available |
not available |
2 |
User mode IP/backtrace samples only |
not available |
available |
not available |
not available |
1 |
Kernel and user mode IP/backtrace samples |
not available |
available |
not available |
not available |
0, -1 |
Kernel and user mode IP/backtrace samples |
Kernel and user mode IP/backtrace samples |
available |
available |
hardware and OS events |
Kernel Version
To support the CPU profiling features utilized by Nsight Systems, the kernel version must be greater than or equal to v4.3. RedHat has backported the required features to the v3.10.0-693 kernel. RedHat distros and their derivatives (e.g. CentOS) require a 3.10.0-693 or later kernel. Use the
uname -rcommand to check the kernel’s version.perf_event_open syscall
The perf_event_open syscall needs to be available. When running within a Docker container, the default seccomp settings will normally block the perf_event_open syscall. To workaround this issue, use the Docker
run --privilegedswitch when launching the docker or modify the docker’s seccomp settings. Some VMs (virtual machines), e.g. AWS, may also block the perf_event_open syscall.Sampling Trigger
In some rare case, a sampling trigger is not available. The sampling trigger is either a hardware or software event that causes a sample to be collected. Some VMs block hardware events from being accessed and therefore, prevent hardware events from being used as sampling triggers. In those cases, Nsight Systems will fall back to using a software trigger if possible.
Checking Your Target System
Use the
nsys status --environmentcommand to check if a system meets the Nsight Systems CPU profiling requirements. Example output from this command is shown below. Note that this command does not check for Linux capability overrides - i.e. if the user or executable files have CAP_SYS_ADMIN or CAP_PERFMON capability. Also, note that this command does not indicate if system-wide mode can be used.
Configuring a CPU Profiling Collection
When configuring Nsight Systems for CPU Profiling from the CLI, use some or all
of the following options: --sample, --cpuctxsw, --event-sample,
--backtrace, --cpu-core-events, --event-sampling-frequency,
--os-events, --samples-per-backtrace, and --sampling-period.
Details about these options, including examples can be found at Profiling from the CLI.
When configuring from the GUI, the following options are available:
The configuration used during CPU profiling is documented in the Analysis Summary:
As well as in the Diagnosics Summary:
Visualizing CPU Profiling Results
Here are example screenshots visualizing CPU profiling results. For details about navigating the Timeline View and the backtraces, see the section on Timeline View in the Reading Your Report in the GUI section of the User Guide.
Example of CPU IP/Backtrace Data
In the timeline, yellow-orange marks can be found under each thread’s timeline that indicate the moment an IP / backtrace sample was collected on that thread (e.g. see the yellow-orange marks in the Specific Samples box above). Hovering the cursor over a mark will cause a tooltip to display the backtrace for that sample.
Below the Timeline is a drop-down list with multiple options including Events View, Top-Down View, Bottom-Up View, and Flat View. All four of these views can be used to view CPU IP / back trace sampling data.
Example of Event Sampling
Event sampling samples hardware or software event counts during a collection and then graphs those events as rates on the Timeline. The above screenshot shows four hardware events. Core and cache events are graphed under the associated CPU row (see the red box in the screenshot) while uncore and OS events are graphed in their own row (see the green box in the screenshot). Hovering the cursor over an event sampling row in the timeline shows the event’s rate at that moment.
Common Issues
Reducing Overhead Caused By Sampling
There are several ways to reduce overhead caused by sampling.
disable sampling (i.e. use the
--sampling=noneswitch)increase the sampling period (i.e. reduce the sampling rate) using the
--sampling-periodswitchstop collecting backtraces (i.e. use the
--backtrace=noneswitch) or collect more efficient backtraces - if available, use the--backtrace=lbrswitch.reduce the number of backtraces collected per sample. See documentation for the
--samples-per-backtraceswitch.
Throttling
The Linux operating system enforces a maximum time to handle sampling interrupts. This means that if collecting samples takes more than a specified amount of time, the OS will throttle (i.e slow down) the sampling rate to prevent the perf subsystem from causing too much overhead. When this occurs, sampling data may become irregular even though the thread is very busy.
The above screenshot shows a case where CPU IP / backtrace sampling was throttled during a collection. Note the irregular intervals of sampling tickmarks on the thread timeline. The number of times a collection throttled is provided in the Nsight Systems GUI’s Diagnostics messages. If a collection throttles frequently (e.g. 1000s of times), increasing the sampling period should help reduce throttling.
Note
When throttling occurs, the OS sets a new (lower) maximum sampling rate in the procfs. This value must be reset before the sampling rate can be increased again. Use the following command to reset the OS’ max sampling rate
echo '100000' | sudo tee /proc/sys/kernel/perf_event_max_sample_rateSample intervals are irregular
My samples are not periodic - why? My samples are clumped up - why? There are gaps in between the samples - why? Likely reasons:
Throttling, as described above
The paranoid level is set to 2. If the paranoid level is set to 2, anytime the workload makes a system call and spends time executing kernel mode code, samples will not be collected and there will be gaps in the sampling data.
The sampling trigger itself is not periodic. If the trigger event is not periodic, for example, the Instructions Retired. event, sample collection will primarily occur when cache misses are occurring.
No CPU profiling data is collected
There are a few common issues that cause CPU profiling data to not be collected
System requirements are not met. Check your system settings with the
nsys status --environmentcommand and see the System Requirements section above.I profiled my workload in a Docker container but no sampling data was collected. By default, Docker containers prevent the perf_event_open syscall from being utilized. To override this behavior, launch the Docker with the
--privilegedswitch or modify the Docker’s seccompsettings.I profiled my workload in a Docker container running Ubuntu 20+ running on top of a host system running CentOS with a kernel version < 3.10.0-693. The
nsys status --environmentcommand indicated that CPU profiling was supported. The host OS kernel version determines if CPU profiling is allowed and a CentOS host with a version < 3.10.0-693 is too old. In this case, thensys status --environmentcommand is incorrect.
NVIDIA Video Profiling
NVIDIA Video Hardware Profiling
Limitations/Requirements
NVIDIA Video Hardware profiling requires:
Linux (x86_64 or Arm) and Windows (x86_64)
Only covers desktop platforms running ResMan kernel driver
Driver version >= 535
GPU architecture Turing+
No NVIDIA Video Hardware profiling for:
Mobile platforms
Driver version < 535
GPU architecture < Turing
GSP is enabled and Driver < 545.31
MIG is enabled
Confidential computing is enabled
vGPU
To learn more about GSP and on which GPUs it’s enabled by default, see the following link.
To turn off GSP permanently:
sudo su -c 'echo options nvidia NVreg_EnableGpuFirmware=0 > /etc/modprobe.d/nvidia-gsp.conf'
sudo update-initramfs -u # for Ubuntu-based systems
Then reboot.
Alternatively if you do not wish to reboot, this will disable until the next reboot:
sudo rmmod nvidia_uvm nvidia_drm nvidia_modeset nvidia && \
sudo insmod /lib/modules/$(uname -r)/updates/dkms/nvidia.ko NVreg_EnableGpuFirmware=0
for i in $(seq 0 7); do sudo nvidia-smi -i $i -pm ENABLED; done
Running from the CLI
The feature is enabled through the --gpu-video-device option. It is available
from the nsys profile, nsys launch and nsys start commands.
The option behaves exactly like --gpu-metrics-device and accepts the
following arguments:
--gpu-video-device help- List supported devices and their IDs, List unsupported devices (if any) and the reason.--gpu-video-device none- Turn the feature off.--gpu-video-device all- Turn the feature on on all supported devices. An error is returned if no devices support the feature.--gpu-video-device <id1,id2,...>- Turn the feature on the specified devices. The ID corresponds to whathelpreturns. An error is returned if the ID is invalid.
Example:
$ nsys profile --gpu-video-device help
Possible --gpu-video-device values are:
0: NVIDIA GeForce RTX 3070 PCI[0000:65:00.0]
all: Select all supported GPUs
none: Disable GPU video accelerator tracing [Default]
Some GPUs don't support video accelerator tracing:
Quadro P620 PCI[0000:04:00.0] (reason = Arch Pascal < Turing)
See the user guide: https://docs.nvidia.com/nsight-systems/UserGuide/index.html
Note that this is a system-wide feature; i.e., it doesn’t require a program to be launched.
NVIDIA Video Codec SDK Trace
Nsight Systems for x86 Linux and Windows targets can trace calls from the NV Video Codec SDK. This software trace can be launched from the GUI or using the --trace nvvideo from the CLI
On the timeline, calls on the CPU to the NV Encoder API and NV Decoder API will be shown.
NV Encoder API Functions Traced by Default
NvEncodeAPICreateInstance
nvEncOpenEncodeSession
nvEncGetEncodeGUIDCount
nvEncGetEncodeGUIDs
nvEncGetEncodeProfileGUIDCount
nvEncGetEncodeProfileGUIDs
nvEncGetInputFormatCount
nvEncGetInputFormats
nvEncGetEncodeCaps
nvEncGetEncodePresetCount
nvEncGetEncodePresetGUIDs
nvEncGetEncodePresetConfig
nvEncGetEncodePresetConfigEx
nvEncInitializeEncoder
nvEncCreateInputBuffer
nvEncDestroyInputBuffer
nvEncCreateBitstreamBuffer
nvEncDestroyBitstreamBuffer
nvEncEncodePicture
nvEncLockBitstream
nvEncUnlockBitstream
nvEncLockInputBuffer
nvEncUnlockInputBuffer
nvEncGetEncodeStats
nvEndGetSequenceParams
nvEncRegisterAsyncEvent
nvEncUnregisterAsyncEvent
nvEncMapInputResource
nvEncUnmapInputResource
nvEncDestroyEncoder
nvEncInvalidateRefFrames
nvEncOpenEncodeSessionEx
nvEncRegisterResource
nvEncUnregisterResource
nvEncReconfigureEncoder
nvEncCreateMVBuffer
nvEncDestroyMVBuffer
nvEncRunMotionEstimationOnly
nvEncGetLastErrorString
nvEncSetIOCudaStreams
nvEncGetSequenceParamEx
NV Decoder API Functions Traced by Default
cuvidCreateVideoSource
cuvidCreateVideoSourceW
cuvidDestroyVideoSource
cuvidSetVideoSourceState
cudaVideoState
cuvidGetSourceVideoFormat
cuvidGetSourceAudioFormat
cuvidCreateVideoParser
cuvidParseVideoData
cuvidDestroyVideoParser
cuvidCreateDecoder
cuvidDestroyDecoder
cuvidDecodePicture
cuvidGetDecodeStatus
cuvidReconfigureDecoder
cuvidMapVideoFrame
cuvidUnmapVideoFrame
cuvidMapVideoFrame64
cuvidUnmapVideoFrame64
cuvidCtxLockCreate
cuvidCtxLockDestroy
cuvidCtxLock
cuvidCtxUnlock
NV JPEG API Functions Traced by Default
nvjpegBufferDeviceCreate
nvjpegBufferDeviceDestroy
nvjpegBufferDeviceRetrieve
nvjpegBufferPinnedCreate
nvjpegBufferPinnedDestroy
nvjpegBufferPinnedRetrieve
nvjpegCreate
nvjpegCreateEx
nvjpegCreateSimple
nvjpegDecode
nvjpegDecodeBatched
nvjpegDecodeBatchedEx
nvjpegDecodeBatchedInitialize
nvjpegDecodeBatchedPreAllocate
nvjpegDecodeBatchedSupported
nvjpegDecodeBatchedSupportedEx
nvjpegDecodeJpeg
nvjpegDecodeJpegDevice
nvjpegDecodeJpegHost
nvjpegDecodeJpegTransferToDevice
nvjpegDecodeParamsCreate
nvjpegDecodeParamsDestroy
nvjpegDecodeParamsSetAllowCMYK
nvjpegDecodeParamsSetOutputFormat
nvjpegDecodeParamsSetROI
nvjpegDecodeParamsSetScaleFactor
nvjpegDecoderCreate
nvjpegDecoderDestroy
nvjpegDecoderJpegSupported
nvjpegDecoderStateCreate
nvjpegDestroy
nvjpegEncodeGetBufferSize
nvjpegEncodeImage
nvjpegEncodeRetrieveBitstream
nvjpegEncodeRetrieveBitstreamDevice
nvjpegEncoderParamsCopyHuffmanTables
nvjpegEncoderParamsCopyMetadata
nvjpegEncoderParamsCopyQuantizationTables
nvjpegEncoderParamsCreate
nvjpegEncoderParamsDestroy
nvjpegEncoderParamsSetEncoding
nvjpegEncoderParamsSetOptimizedHuffman
nvjpegEncoderParamsSetQuality
nvjpegEncoderParamsSetSamplingFactors
nvjpegEncoderStateCreate
nvjpegEncoderStateDestroy
nvjpegEncodeYUV,(nvjpegHandle_t handle
nvjpegGetCudartProperty
nvjpegGetDeviceMemoryPadding
nvjpegGetImageInfo
nvjpegGetPinnedMemoryPadding
nvjpegGetProperty
nvjpegJpegStateCreate
nvjpegJpegStateDestroy
nvjpegJpegStreamCreate
nvjpegJpegStreamDestroy
nvjpegJpegStreamGetChromaSubsampling
nvjpegJpegStreamGetComponentDimensions
nvjpegJpegStreamGetComponentsNum
nvjpegJpegStreamGetFrameDimensions
nvjpegJpegStreamGetJpegEncoding
nvjpegJpegStreamParse
nvjpegJpegStreamParseHeader
nvjpegSetDeviceMemoryPadding
nvjpegSetPinnedMemoryPadding
nvjpegStateAttachDeviceBuffer
nvjpegStateAttachPinnedBuffer
Network Communication Profiling
Nsight Systems can be used to profiles several popular network communication protocols. To enable this, please select the Communication profiling options dropdown.
Then select the libraries you would like to trace:
The corresponding Nsight Systems CLI --trace|-t options are mpi,
oshmem and ucx. For multi-node runs, please refer to section on
Handling Application Launchers in the Profiling From the CLI topic.
MPI API Trace
Nsight Systems has built-in API trace support for Open MPI and MPICH
based MPI implementations via --trace=mpi or by selecting the MPI checkbox
under Network profiling options. If the auto-detection of the MPI
implementation fails, it is possible to specify
it via --mpi-impl=[openmpi|mpich] or the respective checkbox in the GUI.
Nsight Systems will trace a subset of the MPI API, including blocking and non-blocking point-to-point and collective communications as well as MPI one-sided communications, file I/O, and pack operations (see MPI functions traced).
If you require more control over the list of traced APIs or if you are using a different MPI implementation, you can use the NVTX wrappers for MPI on GitHub. Choose an NVTX domain name other than “MPI”, since it is filtered out by Nsight Systems when MPI tracing is not enabled. Use the NVTX-instrumented MPI wrapper library as follows:
nsys profile -e LD_PRELOAD=${PATH_TO_YOUR_NVTX_MPI_LIB} --trace=nvtx
Note
If not all ranks are traced, NSYS_MPI_STORE_TEAMS_PER_RANK has to be set to 1. If communicator tracking is still causing issues, it can be disabled by setting NSYS_MPI_DISABLE_COMMUNICATOR_TRACKING=1.
MPI Communication Parameters
Nsight Systems can get additional information about MPI communication parameters. Currently, the parameters are only visible in the mouseover tooltips or in the eventlog. This means that the data is only available via the GUI. Future versions of the tool will export this information into the SQLite data files for postrun analysis.
In order to fully interpret MPI communications, data for all ranks associated with a communication operation must be loaded into Nsight Systems.
Here is an example of MPI_COMM_WORLD data. This does not require any additional team data, since local rank is the same as global rank.
(Screenshot shows communication parameters for an MPI_Bcast call on rank 3)
When not all processes that are involved in an MPI communication are loaded into Nsight Systems the following information is available.
Right-hand screenshot shows a reused communicator handle (last number increased).
Encoding: MPI_COMM[*team size*]*global-group-root-rank*.*group-ID*
When all reports are loaded into Nsight Systems:
World rank is shown in addition to group-local rank “(world rank X)”
Encoding: MPI_COMM[*team size*]{rank0, rank1, …}
At most 8 ranks are shown (the numbers represent world ranks, the position in the list is the group-local rank)
MPI functions traced
MPI_Init[_thread], MPI_Finalize
MPI_Send, MPI_{B,S,R}send, MPI_Recv, MPI_Mrecv
MPI_Sendrecv[_replace]
MPI_Barrier, MPI_Bcast
MPI_Scatter[v], MPI_Gather[v]
MPI_Allgather[v], MPI_Alltoall[{v,w}]
MPI_Allreduce, MPI_Reduce[_{scatter,scatter_block,local}]
MPI_Scan, MPI_Exscan
MPI_Isend, MPI_I{b,s,r}send, MPI_I[m]recv
MPI_{Send,Bsend,Ssend,Rsend,Recv}_init
MPI_Start[all]
MPI_Ibarrier, MPI_Ibcast
MPI_Iscatter[v], MPI_Igather[v]
MPI_Iallgather[v], MPI_Ialltoall[{v,w}]
MPI_Iallreduce, MPI_Ireduce[{scatter,scatter_block}]
MPI_I[ex]scan
MPI_Wait[{all,any,some}]
MPI_Put, MPI_Rput, MPI_Get, MPI_Rget
MPI_Accumulate, MPI_Raccumulate
MPI_Get_accumulate, MPI_Rget_accumulate
MPI_Fetch_and_op, MPI_Compare_and_swap
MPI_Win_allocate[_shared]
MPI_Win_create[_dynamic]
MPI_Win_{attach, detach}
MPI_Win_free
MPI_Win_fence
MPI_Win_{start, complete, post, wait}
MPI_Win_[un]lock[_all]
MPI_Win_flush[_local][_all]
MPI_Win_sync
MPI_File_{open,close,delete,sync}
MPI_File_{read,write}[_{all,all_begin,all_end}]
MPI_File_{read,write}_at[_{all,all_begin,all_end}]
MPI_File_{read,write}_shared
MPI_File_{read,write}_ordered[_{begin,end}]
MPI_File_i{read,write}[_{all,at,at_all,shared}]
MPI_File_set_{size,view,info}
MPI_File_get_{size,view,info,group,amode}
MPI_File_preallocate
MPI_Pack[_external]
MPI_Unpack[_external]
OpenSHMEM Library Trace
If OpenSHMEM library trace is selected Nsight Systems will trace the subset of OpenSHMEM API functions that are most likely be involved in performance bottlenecks. To keep overhead low Nsight Systems does not trace all functions.
OpenSHMEM 1.5 Functions Not Traced
shmem_my_pe
shmem_n_pes
shmem_global_exit
shmem_pe_accessible
shmem_addr_accessible
shmem_ctx_{create,destroy,get_team}
shmem_global_exit
shmem_info_get_{version,name}
shmem_{my_pe,n_pes,pe_accessible,ptr}
shmem_query_thread
shmem_team_{create_ctx,destroy}
shmem_team_get_config
shmem_team_{my_pe,n_pes,translate_pe}
shmem_team_split_{2d,strided}
shmem_test*
UCX API Trace
If UCX API trace is selected Nsight Systems will trace the subset of functions of the UCX protocol layer UCP that are most likely be involved in performance bottlenecks. To keep overhead low Nsight Systems does not trace all functions.
The following environment variables control what is recorded:
NSYS_UCP_COMM_SUBMIT: (enabled by default) If set to0, UCP communication submission calls are not recorded any more. These calls are usually short, because the communication itself is handled in a worker thread.NSYS_UCP_COMM_PROGRESS: (enabled by default) If set to0, tracking of (process-local) UCP communication progress is disabled. The progress tracking uses UCP completion callbacks.NSYS_UCP_COMM_PARAMS: (enabled by default) If set to0, UCP communication parameters (tag, remote worker UID, packed message size, buffer address) will not be recorded. Recording the remote worker UID requires UCX >= 1.12.0. Recording the packed message size requires UCX >= 1.14.0.
UCX functions traced
ucp_am_send_nb[x]
ucp_am_recv_data_nbx
ucp_am_data_release
ucp_atomic_{add{32,64},cswap{32,64},fadd{32,64},swap{32,64}}
ucp_atomic_{post,fetch_nb,op_nbx}
ucp_cleanup
ucp_config_{modify,read,release}
ucp_disconnect_nb
ucp_dt_{create_generic,destroy}
ucp_ep_{create,destroy,modify_nb,close_nbx}
ucp_ep_flush[{_nb,_nbx}]
ucp_listener_{create,destroy,query,reject}
ucp_mem_{advise,map,unmap,query}
ucp_{put,get}[_nbi]
ucp_{put,get}_nb[x]
ucp_request_{alloc,cancel,is_completed}
ucp_rkey_{buffer_release,destroy,pack,ptr}
ucp_stream_data_release
ucp_stream_recv_data_nb
ucp_stream_{send,recv}_nb[x]
ucp_stream_worker_poll
ucp_tag_msg_recv_nb[x]
ucp_tag_{send,recv}_nbr
ucp_tag_{send,recv}_nb[x]
ucp_tag_send_sync_nb[x]
ucp_worker_{create,destroy,get_address,get_efd,arm,fence,wait,signal,wait_mem}
ucp_worker_flush[{_nb,_nbx}]
ucp_worker_set_am_{handler,recv_handler}
UCX Functions Not Traced:
ucp_config_print
ucp_conn_request_query
ucp_context_{query,print_info}
ucp_get_version[_string]
ucp_ep_{close_nb,print_info,query,rkey_unpack}
ucp_mem_print_info
ucp_request_{check_status,free,query,release,test}
ucp_stream_recv_request_test
ucp_tag_probe_nb
ucp_tag_recv_request_test
ucp_worker_{address_query,print_info,progress,query,release_address}
Additional API functions from other UCX layers may be added in a future version of the product.
NVIDIA NVSHMEM and NCCL Trace
The NVIDIA network communication libraries NVSHMEM and NCCL have been instrumented using NVTX annotations. To enable tracing these libraries in Nsight Systems, turn on NVTX tracing in the GUI or CLI. To enable the NVTX instrumentation of the NVSHMEM library, make sure that the environment variable NVSHMEM_NVTX is set properly, e.g. NVSHMEM_NVTX=common.
NIC Metric Sampling
Overview
NVIDIA ConnectX smart network interface cards (smart NICs) offer advanced hardware offloads and accelerations for network operations. Viewing smart NICs metrics, on Nsight Systems timeline, enables developers to better understand their application’s network usage. Developers can use this information to optimize the application’s performance.
Limitations/Requirements
NIC metric sampling supports NVIDIA ConnectX boards starting with ConnectX 5
NIC metric sampling is supported on Linux x86_64 and Arm Server (SBSA) machines only, having minimum Linux kernel 4.12 and minimum MLNX_OFED 4.1. You can download the latest and archived versions of the MLX_OFED driver from the MLNX_OFED Download Center. If collecting NIC metrics within a container, make sure that the container has access to the driver on the host machine. To check manually if OFED is installed and get its version you can run:
/usr/bin/ofed_infocat /sys/module/"$(cat /proc/modules | grep -o -E "^mlx._core")"/version
It might take several hundreds of milliseconds for the NIC metrics collector to initialize. Thus, we recommend that the profiling session should last at least 1 second when collecting NIC metrics.
To check if the target system meets the requirements for NIC metrics collection you can run nsys status --network.
Collecting NIC Metrics Using the Command Line
To collect NIC performance metrics, using Nsight Systems CLI, add the --nic-metrics command line switch:
nsys profile --nic-metrics=true my_app
Available Metrics
Bytes sent - Number of bytes sent through all NIC ports.
Bytes received - Number of bytes received by all NIC ports.
Average sent packet size - Average byte size of packets sent through all NIC ports.
Average received packet size - Average byte size of packets received by all NIC ports.
CNPs sent - Number of congestion notification packets sent by the NIC.
CNPs received - Number of congestion notification packets received and handled by the NIC.
Send waits - The number of ticks during which ports had data to transmit but no data was sent during the entire tick (either because of insufficient credits or because of lack of arbitration)
Note: Each one of the mentioned metrics is shown only if it has non-zero value during profiling.
Usage Examples
The
Bytes sent/secand theBytes received/secmetrics enables identifying idle and busy NIC times.Developers may shift network operations from busy to idle times to reduce network congestion and latency.
Developers can use idle NIC times to send additional data without reducing application performance.
CNPs (congestion notification packets) received/sent and Send waits metrics may explain network latencies. A developer seeing the time periods when the network was congested may rewrite his algorithm to avoid the observed congestions.
InfiniBand Switch Metric Sampling
NVIDIA Quantum InfiniBand switches offer high-bandwidth, low-latency communication. Viewing switch metrics, on Nsight Systems timeline, enables developers to better understand their application’s network usage. Developers can use this information to optimize the application’s performance.
Limitations/Requirements
IB switch metric sampling supports all NVIDIA Quantum switches. The user needs to have permission to query the InfiniBand switch metrics.
To check if the current user has permissions to query the InfiniBand switch metrics, check that the user have permission to access /dev/infiniband/umad*
To give user permissions to query InfiniBand switch metrics on RedHat systems, follow the directions at RedHat Solutions.
To collect InfiniBand switch performance metric, using Nsight Systems CLI, add
the --ib-switch-metrics-device command line switch, followed by a comma
separated list of InfiniBand switch GUIDs. For example:
nsys profile --ib-switch-metrics-device=<IB switch GUID> my_app
To get a list of InfiniBand switches, reachable by a given NIC, use:
sudo ibswitches -C <nic name>
Available Metrics
Bytes sent - Number of bytes sent through all switch ports
Bytes received - Number of bytes received by all switch ports
- Send waits - The number of ticks during which switch ports, selected by
PortSelect, had data to transmit but no data was sent during the entire tick (either because of insufficient credits or of lack of arbitration)
Average sent packet size - Average sent InfiniBand packet size
Average received packet size - Average received InfiniBand packet size
InfiniBand Switch Congestion Events
Overview
NVIDIA Quantum InfiniBand switches offer high-bandwidth, low-latency communication.
When a switch egress port is congested, packets wait in the egress port queue before being sent out of the switch. This increases the latency of these packets.
Nsight Systems Workstation Edition gives you the ability to view when switch egress ports are congested on the Nsight Systems timeline. This enables developers to better understand latencies that are caused by the application’s network usage. Developers can use this information to optimize the application’s performance.
Limitations/Requirements
IB switch congestion events support requires:
Quantum 2 switch or newer
Having firmware version 31.2012.1068 or higher
User need to have permission to send management datagrams
To get a list of InfiniBand switches, reachable by a given NIC, use:
sudo ibswitches -C <nic name>
To check if the current user has permissions to send management datagrams, check
that the user has permission to access /dev/umad. To give user permissions
to query InfiniBand switch metrics on RedHat systems, follow the directions given
at RedHat Solutions.
Using the Command Line
To collect InfiniBand switch congestion events, using Nsight Systems CLI, add the following command line switches:
ib-switch-congestion-deviceThis should be followed by a comma separated list of InfiniBand switch GUIDs, from which congestion events will be collected.ib-switch-congestion-nic-deviceThis should be followed by the name of the NIC (HCA) through which InfiniBand switches will be accessed. The profiled InfiniBand switches should be reachable by this NIC.ib-switch-congestion-percentThis defines the percent of InfiniBand switch congestion events to be collected. This option enables reducing the network bandwidth consumed by reporting congestion events. Values are in the [1,100] range.ib-switch-congestion-threshold-highThis defines the high threshold for InfiniBand switch egress port queue size. When a packet enters an InfiniBand switch, its data is stored at an ingress port buffer. A pointer to the packet’s data is inserted into the egress port’s queue, from which the packet will be exiting the switch. At that point, the threshold given by this command switch is compared to the egress queue data size. If the queue data size exceeds the threshold, a congestion event is reported. The threshold is given in percent of the ingress port size. An egress port queue can point to data coming from multiple ingress port buffers, therefore the threshold can be bigger than 100%. Values are in the [1,1023] range
InfiniBand Network Information
Overview
By default, Nsight Systems displays low-level identifiers like LIDs (Local Identifiers) and GUIDs (Globally Unique Identifiers). Instead, Nsight Systems can leverage InfiniBand network information to display the actual names of nodes and switches. This makes the Nsight Systems reports much more intuitive and easier to understand at a glance.
InfiniBand network information discovery is done using the ibdiagnet utility. Either:
Run ibdiagnet and store the generated network information files to be later used by Nsight Systems. - This method is useful for large networks, where ibdiagnet’s network discovery time may be long, and for networks where only administrators have permissions to query the network information.
A user can ask Nsight Systems to run ibdiagnet to collect the network information during the profiling session. - This method is useful for small networks.
Limitations/Requirements
The user needs to have permission to send MADs (management datagrams). To check if you have permission to send MADs, check if you can access the /dev/infiniband/umad* files. To give user permissions to send MADs on RedHat systems, follow the directions at RedHat Solutions.
Relevant Switches
The following Nsight Systems command line switches enable collecting InfiniBand network information:
ib-net-info-devicesThis should be followed by a comma separated list of NIC names, from which ibdiagnet will run network discovery. The results of the network discovery will be automatically loaded into Nsight Systems.
ib-net-info-filesThis should be followed by a comma separated list of pre-generated ibdiagnet db_csv file paths, which Nsight Systems will read.
ib-net-info-outputThis should be followed by a path of a directory into which Nsight Systems will store the ibdiagnet network discovery data. These files will be used by the
ib-net-info-devicescommand line switch. This command line switch can only be used together with theib-net-info-devicescommand line switch.
The above image displays a congestion event. InfiniBand network information is used for displaying node and switch names instead of LIDs.
Amazon AWS EFA Metrics (preview)
Nsight Systems can now periodically sample performance counters for AWS Elastic Fabric Adapters (EFAs) and plot it on the timeline in the GUI. This enables developers to analyze how network communications may be involved with the critical path of their multi-node application. Created in collaboration with AWS, this plugin will work on AWS EC2 NVIDIA GPU accelerated compute instances .
To enable the AWS EFA metrics add the following option to the nsys
profile or start commands:
--enable efa_metrics[,arg1[=value1],arg2[=value2], ...]
There are no spaces following efa_metrics plugin name. It is followed by a
comma separated list of arguments or argument=value pairs. Arguments with spaces
should be enclosed in double quotes.
Supported arguments are:
Name |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|
|
true, false |
false |
Sample Infiniband non-RDMA counters |
|
<path> |
/sys/class/infiniband |
Root directory for EFA counters sysfs |
|
true, false |
false |
Sample Infiniband WorkRequest counters |
|
true, false |
false |
Sample error counters |
|
integer, a negative value means 1/F frequency |
10 |
Target sample frequency in hertz |
|
throughput, delta, total |
throughput |
Report sampled counters as a value per second, delta since previous sample, or an accumulated sum. |
|
true, false |
false |
Sample packet counters |
Usage Examples
nsys profile --enable efa_metrics ...Sample all EFA adapters, display as bytes per second.
nsys profile --enable efa_metrics,-packets,-errors,-efa-non-rdma ...Sample all available EFA adapter counters.
nsys profile --enable efa_metrics,-mode=total ...Sample all EFA adapters, display as total value sum since profiling start.
nsys profile --enable efa_metrics,-efa-counters-sysfs="/mnt/nv/sys", ...Look for EFA counters in a different sysfs directory. Useful in some k8s environments.
This collector is the first use case for the Nsight Systems Plugins (Preview) system.
Network interface metrics (preview)
Nsight Systems can now periodically sample performance counters for network interface devices and plot them on the timeline in the GUI.
To enable the network devices metrics add the following option to the nsys
profile or start commands:
--enable network_interface[,arg1[=value1],arg2[=value2], ...]
There are no spaces following network_interface plugin name. It is followed by a
comma separated list of arguments or argument=value pairs. Arguments with spaces
should be enclosed in double quotes.
Supported arguments are:
Short name |
Long name |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|---|
|
|
integer |
100000 |
Sampling interval in microseconds |
|
|
regular expression |
“.+” (and filtering for physical devices) |
Device(s) to sample |
|
|
regular expression |
“.*_bytes” |
Metric(s) to sample |
|
|
Print help message |
Usage Examples
nsys profile --enable network_interface ...Sample bytes metrics for all physical network devices every 100ms.
nsys profile --enable network_interface,-dall ...Sample bytes metrics for all network devices every 100ms.
nsys profile --enable network_interface,-i10000,-dall,-m".+"Sample all metrics, for all network devices, every 10ms.
For general information on Nsight Systems plugins please refer to Nsight Systems Plugins (Preview) system.
Python Profiling
Nsight Systems has several features that have been added in the last few years to enhance users optimizing their python code.
Note
You may find that all of your python application output comes at the end of the run instead of as events happen.
Python will change the buffering of stdout depending on whether it points to
a tty or something else. Nsight Systems redirects the application stdout to
a pipe to demultiplex stdout to both a file and the terminal. As a side
effect, it makes Python change stdout buffering from line-buffered to
page-buffered. You can use python -u option or the PYTHONUNBUFFERED
environment variable to override this behavior.
Python Backtrace Sampling
Nsight Systems for Arm server (SBSA) platforms, x86 Linux and Windows targets, is capable of periodically capturing Python backtrace information. This functionality is available when tracing Python interpreters of version 3.9 or later. Capturing python backtrace is done in periodic samples, in a selected frequency ranging from 1Hz - 2KHz with a default value of 1KHz. Note that this feature provides meaningful backtraces for Python processes, when profiling Python-only workflows, consider disabling the CPU sampling option to reduce overhead.
To enable Python backtrace sampling from Nsight Systems:
CLI — Set --python-sampling=true and use the --python-sampling-frequency option to set the sampling rate.
GUI — Select the Collect Python backtrace samples checkbox.
Example screenshot:
Python Functions Trace
Nsight Systems for Arm server (SBSA) platforms, x86 Linux and Windows targets, is capable of using NVTX to annotate Python functions.
The Python source code does not require any changes. This feature requires CPython interpreter, release 3.8 or later.
The annotations are configured in a JSON file. An example file is located in Nsight Systems installation folder in <target-platform-folder>/PythonFunctionsTrace/annotations.json.
Notes:
Annotating function from module
__main__is not supported.
To enable Python functions trace from Nsight Systems:
CLI — Set --python-functions-trace=<json_file>.
GUI — Select the Python Functions trace checkbox and specify the JSON file.
Example screenshot:
Python GIL Tracing
Nsight Systems for Arm server (SBSA) platforms, x86 Linux and Windows targets, is capable of tracing when Python threads are waiting to hold and holding the GIL (Global Interpreter Lock).
The Python source code does not require any changes. This feature requires CPython interpreter, release 3.9 or later.
CLI — Set --trace=python-gil.
GUI — Select the Trace GIL checkbox under Python profiling options.
Example screenshot:
Profiling Embedded Virtual Machines
Nsight Systems and DRIVE Hypervisor support periodic CPU sampling with call stacks. It works both on DRIVE Linux and QNX.
The call stacks are collected using frame pointers. The Linux kernel, QNX kernel, and user space libraries provided by NVIDIA are compiled with frame pointers. To ensure correct call stacks, we recommend compiling all application code with frame pointer support, using -fno-omit-frame-pointer with GCC, Clang, and QCC.
This is an experimental feature and is expected to change in the future.
The symbols can be resolved both for user space code, and for kernel space code:
In the user space, the VMProfiler events are matched with the CPU thread state trace coming from Linux Perf and QNX Tracelogger. After that, Nsight Systems can know the module filename, and can resolve symbols directly from these files if they are unstripped, or by looking up additional files with symbols. See more details below.
In the kernel space (Linux kernel, QNX kernel, and additional service VMs), the symbols are resolved using the ELF file with symbols specified.
kernel_symbols.jsoninput file specifies the location of this ELF file.
Please follow the steps below to learn how to:
Flash the devkit (these steps are given just as an example, the exact steps might differ in your case).
Copy the necessary files: pct.json, eventlib schema files, and kernel_symbols.json.
Compose kernel_symbols.json to allow resolving symbols in the Linux kernel, QNX kernel, and additional service VMs.
See example CLI commands to collect data.
Known issues:
At the moment, this feature is not compatible with standard CPU sampling on Linux and QNX. Please disable it with
-s noneor--sample=none.When enabled together, hypervisor trace plus VMProfiler sampling can write too much data into the same eventlib buffers, and the Nsight Systems agent might not be able to keep up with the rate, losing events. If that happens, please disable hypervisor trace events with
--xhv-trace-events=none.
Flashing DRIVE OS QNX/Linux
Log into the NVIDIA GPU Cloud (NGC):
sudo docker login nvcr.io
Username: ``$oauthtoken``
Password: <NGC API key>
Docker command:
sudo docker run --rm --privileged --net host \
-v /dev/bus/usb:/dev/bus/usb \
-v /tmp:/drive_flashing \
-it <docker image>
<docker image> - docker image link.
Examples:
6.0.8.0 QNX:
sudo docker run --rm --privileged --net host \
-v /dev/bus/usb:/dev/bus/usb \
-v /tmp:/drive_flashing \
-it nvcr.io/{MY_NGC_ORG}/driveos-pdk/drive-agx-orin-qnx-aarch64-pdk-build-x86:6.0.8.0-0003
6.0.9.1 QNX:
sudo docker run --rm --privileged --net host \
-v /dev/bus/usb:/dev/bus/usb \
-v /tmp:/drive_flashing \
-it nvcr.io/{MY_NGC_ORG}/driveos-pdk/drive-agx-orin-qnx-aarch64-pdk-build-x86:6.0.9.1-latest
6.0.8.0 Linux:
sudo docker run --rm --privileged --net host \
-v /dev/bus/usb:/dev/bus/usb \
-v /tmp:/drive_flashing \
-it nvcr.io/{MY_NGC_ORG}/driveos-pdk/drive-agx-orin-linux-aarch64-pdk-build-x86:6.0.8.0-0003
Inside of container, flash with flash.py:
cd /drive
./flash.py <aurix> <board>
<board>- target board base name: ‘p3710’ or ‘p3663’.<aurix>- Aurix serial port, for example: /dev/ttyACM1, /dev/ttyUSB1.
Examples:
Firespray p3710:
./flash.py /dev/ttyACM1 p3710
Drive Orin p3663:
./flash.py /dev/ttyUSB1 p3663
List the available EMMC and UFS partitions:
df -h
Format a power-safe file system partition, and mount it, example for vblk_ufs40:
mkqnx6fs /dev/vblk_ufs40 -q
mount -o rw /dev/vblk_ufs40 /
# df -h
/dev/vblk_ufs40 116G 7.5G 108G 7% /
ifs 16M 16M 0 100% /
ifs 52M 52M 0 100% /
...
Note
For more information about DRIVE OS installation, see the following link: NVIDIA DRIVE OS Documentation (useful pages: DRIVE OS Linux Installation Guide, DRIVE OS QNX Installation Guide).
Create XHV Directory
Inside of container, examples for p3710, QNX/Linux:
QNX:
cd /drive_flashing
mkdir -p xhv/hypervisor/configs/t234ref-release/pct/qnx xhv/schemas
cp -rv /drive/drive-foundation/virtualization/hypervisor/t23x/configs/t234ref-release/pct/p3710-10-a03/qnx/pct.json ./xhv/hypervisor/configs/t234ref-release/pct/qnx/
cp -rv /drive/drive-foundation/schemas/event ./xhv/schemas/
Linux:
cd /drive_flashing
mkdir -p xhv/hypervisor/configs/t234ref-release/pct/linux xhv/schemas
cp -rv /drive/drive-foundation/virtualization/hypervisor/t23x/configs/t234ref-release/pct/p3710-10-a03/linux/pct.json ./xhv/hypervisor/configs/t234ref-release/pct/linux/
cp -rv /drive/drive-foundation/schemas/event ./xhv/schemas/
Example of XHV directory (Linux):
xhv/
├── hypervisor
│ └── configs
│ └── t234ref-release
│ └── pct
│ └── linux
│ └── pct.json
└── schemas
└── event
├── audioserver_events.json
├── bpmp_events.json
├── cem_events.json
├── hv_events.json
├── i2c_events.json
├── Makefile.gen-event-headers.tmk
├── monitor_events.json
├── se_events.json
├── sysmgr_events.json
└── vsc_events.json
Copy XHV directory to target:
scp -r xhv <user>@<target-IP>
eventlib_dump tool (QNX/Linux):
cp -rv /drive/drive-qnx/nvidia-bsp/aarch64le/sbin/eventlib_dump /drive_flashing/
cp -rv /drive/drive-linux/filesystem/contents/bin/eventlib_dump /drive_flashing/
Specific Command Line Options
Option |
Possible Parameters |
Default |
Switch Description |
|---|---|---|---|
|
true, false |
false |
Enable VMProfiler sampling. |
|
< filepath kernel_symbols.json > |
none |
VMProfiler sampling config (optional, for kernel symbols). |
|
< filepath pct.json > |
none |
Collect hypervisor trace. |
|
all, none, core, sched, irq, trap |
all |
HV trace events. |
Examples:
nsys profile --process-scope=process-tree --sample=none --trace=nvtx,osrt,cuda --el1-sampling=true --el1-sampling-config=/root/kernel_symbols.json --xhv-trace=/root/xhv/hypervisor/configs/p3710-10-a01/pct/qnx/pct.json --xhv-trace-events=none sleep 5
nsys profile --process-scope=system-wide --sample=none --el1-sampling=true --el1-sampling-config=/root/kernel_symbols.json --xhv-trace=/root/xhv/hypervisor/configs/p3710-10-a01/pct/qnx/pct.json --xhv-trace-events=none sleep 5
Example screenshot:
Config File (for kernel symbols)
Examples:
QNX, kernel_symbols.json file:
{
"guest_cfg": [
{
"guest_id": 0,
"guest_name": "Guest VM 0",
"symbols": "/root/symbols/procnto-smp-instr-safety.guest_vm.bin.sym"
},
{
"guest_id": 1,
"guest_name": "Update service",
"symbols": "/root/symbols/procnto-smp-instr-safety.update_vm.bin.sym"
},
{
"guest_id": 2,
"guest_name": "Resource Manager Server"
},
{
"guest_id": 3,
"guest_name": "Storage Server"
},
{
"guest_id": 4,
"guest_name": "Ethernet Server"
},
{
"guest_id": 5,
"guest_name": "Debug Server"
}
],
"symbol_files": {
"Sidekick": "/root/symbols/sidekick.unstripped"
}
}
Linux, kernel_symbols.json file:
{
"guest_cfg": [
{
"guest_id": 0,
"guest_name": "Guest VM 0",
"symbols": "/home/nvidia/vmlinux"
},
{
"guest_id": 1,
"guest_name": "Update service"
}
],
"symbol_files": {
}
}
Symbol Files
The list of directories with symbol files:
CLI:
DbgFileSearchPathconfig option, for example:DbgFileSearchPath="/lib:/root/symbols"- list of directories with symbol/debug files. On Linux, the default path is/usr/lib/debug. On QNX, there is no default path.Example:
NSYS_CONFIG_DIRECTIVES='DbgFileSearchPath="/lib:/root/symbols"' nsys profile --process-scope=process-tree --sample=none --el1-sampling=true --el1-sampling-config=/root/kernel_symbols.json --xhv-trace=/root/xhv/hypervisor/configs/p3710-10-a01/pct/qnx/pct.json --xhv-trace-events=none sleep 5
GUI:
Symbol locationbutton.
The search is non-recursive.
There are several ways of searching for symbol files - Nsight Systems tries them sequentially for each target file:
Build-id debug files (CLI only)
<symbol directory>/.build-id/… - directories with debug files (or links to debug files).
Example:
.build-id/ ├── 00 └── 6627b119cc2aee77e10e0535fc243fce8fe66e.debug ├── 01 ├── 3e4007e3cb24359203fc02b63bb90f16db5b23.debug └── fb938bc0f029c41a8e1e88f01f88f75cf3a0d3.debug ...Debuglink files (CLI only)
<symbol directory>/<symbol file> - both filename and CRC from debuglink section must be matched for the symbol file.
File name and build-id (CLI/GUI)
<symbol directory>/<symbol file> - by filename and build-id.
Adding Your Own Collection to a Report
Nsight Systems allows the user to add additional information to a report file for display with other Nsight Systems options.
Nsight Systems Plugins (Preview)
What is a plugin
Nsight Systems plugins are standalone applications that can be profiled along with the main application or without one in a system-wide profiling. The NVTX events emitted by a plugin are displayed in the same timeline as the main application events. Additionally, any stdout and stderr streams are captured the same way as for a target application.
To make a plugin available for profiling, create a directory with a manifest file nsys-plugin.yaml then place it in a “plugins” directory next to Nsight Systems target CLI binary. The manifest file describes the plugin and its configuration.
Manifest file contents
The manifest file is a YAML file with the following required fields:
PluginName: SamplePlugin
ExecutablePath: PluginExecutableRelativeToManifest
Description: This is a sample plugin.
How to launch a plugin
Plugins are supported in nsys profile and nsys start commands. Plugin
processes are launched by Nsight Systems as if it was a target application and
terminated at the end of profiling. It’s possible to launch multiple instances
of the same plugin by using multiple --enable options.
How to pass arguments to a plugin
To pass arguments to a plugin, specify them as a part of --enable option
after plugin name when launching the target application. The arguments should
be separated by commas only (no spaces). Commas can be escaped with a backslash ‘\’.
The backslash itself can be escaped by another backslash ‘\\’. To include
spaces in an argument, enclose the argument in double quotes ‘”’.
See the section on the Amazon AWS Elastic Fabric Adapter (EFA) Network Counters for an example.
Supported platforms
Currently plugins are supported on x86_64 and arm64 Linux.
Sample plugin
You can look at the Nsight Systems installation path, for example /opt/nvidia/nsight-systems/2024.5.1/target-linux-x64, under the directory samples for the source code of a sample plugin.
The NetworkPlugin.cpp source file is the exact source code for the network_interface plugin that ships in binary form with Nsight Systems. Users can modify this plugin or use it as a guide to create their own plugin for profiling their intended source of metrics.
Import NVTXT
ImportNvtxt is an utility which allows conversion of a NVTXT file to a Nsight Systems report file (*.nsys-rep) or to merge it with an existing report file.
Note: NvtxtImport supports custom TimeBase values. Only these values are supported:
Manual — timestamps are set using absolute values.
Relative — timestamps are set using relative values with regards to report file which is being merged with nvtxt file.
ClockMonotonicRaw — timestamps values in nvtxt file are considered to be gathered on the same target as the report file which is to be merged with nvtxt using
clock_gettime(CLOCK_MONOTONIC_RAW, ...)call.- CNTVCT — timestamps values in nvtxt file are considered to be gathered
on the same target as the report file which is to be merged with nvtxt using CNTVCT values.
You can get usage info via help message:
Print help message:
-h [ --help ]
Show information about report file:
--cmd info -i [--input] arg
Create report file from existing nvtxt file:
--cmd create -n [--nvtxt] arg -o [--output] arg [-m [--mode] mode_name mode_args] [--target <Hw:Vm>] [--update_report_time]
Merge nvtxt file to existing report file:
--cmd merge -i [--input] arg -n [--nvtxt] arg -o [--output] arg [-m [--mode] mode_name mode_args] [--target <Hw:Vm>] [--update_report_time]
Modes description:
lerp - Insert with linear interpolation
--mode lerp --ns_a arg --ns_b arg [--nvtxt_a arg --nvtxt_b arg]
lin - insert with linear equation
--mode lin --ns_a arg --freq arg [--nvtxt_a arg]
Modes’ parameters:
ns_a- a nanoseconds valuens_b- a nanoseconds value (greater thanns_a)nvtxt_a- an nvtxt file’s time unit value corresponding tons_ananosecondsnvtxt_b- an nvtxt file’s time unit value corresponding tons_bnanosecondsfreq- the nvtxt file’s timer frequency--target <Hw:Vm>- specify target id, e.g.--target 0:1--update_report_time- prolong report’s profiling session time whilemerging if needed. Without this option all events outside the profiling session time window will be skipped during merging.
Commands
Info
To find out report’s start and end time use info command.
Usage:
ImportNvtxt --cmd info -i [--input] arg
Example:
ImportNvtxt info Report.nsys-rep
Analysis start (ns) 83501026500000
Analysis end (ns) 83506375000000
Create
You can create a report file using existing NVTXT with create command.
Usage:
ImportNvtxt --cmd create -n [--nvtxt] arg -o [--output] arg [-m [--mode] mode_name mode_args]
Available modes are:
lerp — insert with linear interpolation.
lin — insert with linear equation.
Usage for lerp mode is:
--mode lerp --ns_a arg --ns_b arg [--nvtxt_a arg --nvtxt_b arg]
with:
ns_a— a nanoseconds value.ns_b— a nanoseconds value (greater thanns_a).nvtxt_a— an nvtxt file’s time unit value corresponding tons_ananoseconds.nvtxt_b— an nvtxt file’s time unit value corresponding tons_bnanoseconds.
If nvtxt_a and nvtxt_b are not specified, they are respectively set to nvtxt file’s minimum and maximum time value.
Usage for lin mode is:
--mode lin --ns_a arg --freq arg [--nvtxt_a arg]
with:
ns_a— a nanoseconds value.freq— the nvtxt file’s timer frequency.nvtxt_a— an nvtxt file’s time unit value corresponding tons_ananoseconds.
If nvtxt_a is not specified, it is set to nvtxt file’s minimum time value.
Examples:
ImportNvtxt --cmd create -n Sample.nvtxt -o Report.nsys-rep
The output will be a new generated report file which can be opened and viewed by Nsight Systems.
Merge
To merge NVTXT file with an existing report file use merge command.
Usage:
ImportNvtxt --cmd merge -i [--input] arg -n [--nvtxt] arg -o [--output] arg [-m [--mode] mode_name mode_args]
Available modes are:
lerp — insert with linear interpolation.
lin — insert with linear equation.
Usage for lerp mode is:
--mode lerp --ns_a arg --ns_b arg [--nvtxt_a arg --nvtxt_b arg]
with:
ns_a— a nanoseconds value.ns_b— a nanoseconds value (greater thanns_a).nvtxt_a— an nvtxt file’s time unit value corresponding tons_ananoseconds.nvtxt_b— an nvtxt file’s time unit value corresponding tons_bnanoseconds.
If nvtxt_a and nvtxt_b are not specified, they are respectively set to nvtxt file’s minimum and maximum time value.
Usage for lin mode is:
--mode lin --ns_a arg --freq arg [--nvtxt_a arg]
with:
ns_a— a nanoseconds value.freq— the nvtxt file’s timer frequency.nvtxt_a— an nvtxt file’s time unit value corresponding tons_ananoseconds.
If nvtxt_a is not specified, it is set to nvtxt file’s minimum time value.
Time values in <filename.nvtxt> are assumed to be nanoseconds if no mode specified.
Example
ImportNvtxt --cmd merge -i Report.nsys-rep -n Sample.nvtxt -o NewReport.nsys-rep
Reading Your Report in GUI
Generating a New Report
Users can generate a new report by stopping a profiling session. If a profiling session has been canceled, a report will not be generated, and all collected data will be discarded.
A new .nsys-rep file will be created and put into the same directory as the project file (.qdproj).
Opening an Existing Report
An existing .nsys-rep file can be opened using File > Open….
Report Tab
While generating a new report or loading an existing one, a new tab will be created. The most important parts of the report tab are:
View selector — Allows switching between Multi-report view (absent for single reports), Analysis Summary, Timeline View, Diagnostics Summary, and Symbol Resolution Logs views.
Timeline — This is where all charts are displayed.
Function table — Located below the timeline, it displays statistical information about functions in the target application in multiple ways.
Additionally, the following controls are available:
Zoom slider — Allows you to vertically zoom the charts on the timeline.
Analysis Summary View
This view shows a summary of the profiling session. In particular, it is useful to review the project configuration used to generate this report. Information from this view can be selected and copied using the mouse cursor.
Diagnostics Summary View
This view shows important messages. Some of them were generated during the profiling session, while some were added while processing and analyzing data in the report. Messages can be one of the following types:
Informational messages
Warnings
Errors
To draw attention to important diagnostics messages, a summary line is displayed on the timeline view in the top right corner:
Information from this view can be selected and copied using the mouse cursor.
Symbol Resolution Logs View
This view shows all messages related to the process of resolving symbols. It might be useful to debug issues when some of the symbol names in the symbols table of the timeline view are unresolved.
Timeline View
The timeline view consists of two main controls: the timeline at the top, and a bottom pane that contains the events view and the function table. In some cases, when sampling of a process has not been enabled, the function table might be empty and hidden.
The bottom view selector sets the view that is displayed in the bottom pane.
Timeline
Timeline is a versatile control that contains a tree-like hierarchy on the left, a line labels column in the center, and the corresponding charts on the right. The line labels column can be hidden by using the timeline options.
Contents of the hierarchy depend on the project settings used to collect the report. For example, if a certain feature has not been enabled, corresponding rows will not be shown on the timeline.
To generate a timeline screenshot without opening the full GUI, use the command
nsys-ui.exe --screenshot filename.nsys-rep
Hovering over elements in the GUI will cause a tooltip to pop open as appropriate
to give additional information, such as the parameters of that function call or
or the call stack. Tooltips can be copied by hovering and right clicking to bring
up the Copy Tooltip option in the context menu:
Timeline Options
We strongly recommend using the OS/Desktop defaults for size and color, but if you would like to set them for yourself, they are available using the Tools > Options dialog.
The above will change the options globally for this GUI. It’s also possible to change some options for a particular open report. There is an “Options…” button near the View Selector:
This button will show a dialog that allows showing/hiding the following:
correlation arrows;
line labels;
CPU occupancy chart.
By default, the timeline will be based on session time. If you would like to switch to global time, click on the small arrow at the top of the leftmost column to reveal the dropdown shown below:
Events View
The Events View provides a tabular display of the trace events. The view contents can be searched and sorted.
Double-clicking an item in the Events View automatically focuses the Timeline View on the corresponding timeline item.
API calls, GPU executions, and debug markers that occurred within the boundaries of a debug marker are displayed nested to that debug marker. Multiple levels of nesting are supported.
Events view recognizes these types of debug markers:
NVTX
Vulkan VK_EXT_debug_marker markers, VK_EXT_debug_utils labels
PIX events and markers
OpenGL KHR_debug markers
You can copy and paste from the events view by highlighting rows, using Shift or Ctrl to enable multi-select. Right clicking on the selection will give you a copy option.
Pasting into text gives you a tab separated view:
Pasting into spreadsheet properly copies into rows and columns:
Function Table Modes
The function table can work in three modes:
Top-Down View — In this mode, expanding top-level functions provides information about the callee functions. One of the top-level functions is typically the main function of your application, or another entry point defined by the runtime libraries.
Bottom-Up View — This is a reverse of the Top-Down view. On the top level, there are functions directly hit by the sampling profiler. To explore all possible call chains leading to these functions, you need to expand the subtrees of the top-level functions.
Flat View — This view enumerates all functions ever observed by the profiler, even if they have never been directly hit, but just appeared somewhere on the call stack. This view typically provides a high-level overview of which parts of the code are CPU-intensive.
Each of the views helps understand particular performance issues of the application being profiled. For example:
When trying to find specific bottleneck functions that can be optimized, the Bottom-Up view should be used. Typically, the top few functions should be examined. Expand them to understand in which contexts they are being used.
To navigate the call tree of the application and while generally searching for algorithms and parts of the code that consume unexpectedly large amount of CPU time, the Top-Down view should be used.
To quickly assess which parts of the application, or high level parts of an algorithm, consume significant amount of CPU time, use the Flat view.
The Top-Down and Bottom-Up views have Self and Total columns, while the Flat view has a Flat column. It is important to understand the meaning of each of the columns:
Top-Down view
Self column denotes the relative amount of time spent executing instructions of this particular function.
Total column shows how much time has been spent executing this function, including all other functions called from this one. Total values of sibling rows sum up to the Total value of the parent row, or 100% for the top-level rows.
Bottom-Up view
Self column for top-level rows, as in the Top-Down view, shows how much time has been spent directly in this function. Self times of all top-level rows add up to 100%.
Self column for children rows breaks down the value of the parent row based on the various call chains leading to that function. Self times of sibling rows add up to the value of the parent row.
Flat view
Flat column shows how much time this function has been anywhere on the call stack. Values in this column do not add up or have other significant relationships.
Note
If low-impact functions have been filtered out, values may not add up correctly to 100%, or to the value of the parent row. This filtering can be disabled.
Contents of the symbols table is tightly related to the timeline. Users can apply and modify filters on the timeline, and they will affect which information is displayed in the symbols table:
Per-thread filtering — Each thread that has sampling information associated with it has a checkbox next to it on the timeline. Only threads with selected checkboxes are represented in the symbols table.
Time filtering — A time filter can be setup on the timeline by pressing the left mouse button, dragging over a region of interest on the timeline, and then choosing Filter by selection in the dropdown menu. In this case, only sampling information collected during the selected time range will be used to build the symbols table.
Note
If too little sampling data is being used to build the symbols table (for example, when the sampling rate is configured to be low, and a short period of time is used for time-based filtering), the numbers in the symbols table might not be representative or accurate in some cases.
Function Table Notes
Last Branch Records vs Frame Pointers
Two of the mechanisms available for collecting backtraces are Intel Last Branch Records (LBRs) and frame pointers. LBRs are used to trace every branch instruction via a limited set of hardware registers. They can be configured to generate backtraces but have finite depth based on the CPU’s microarchitecture. LBRs are effectively free to collect but may not be as deep as you need in order to fully understand how the workload arrived a specific Instruction Pointer (IP).
Frame pointers only work when a binary is compiled with the -fno-omit-frame-pointer compiler switch. To determine if frame pointers are enabled on an x86_64 binary running on Linux, dump a binary’s assembly code using the objdump -d [binary_file] command and look for this pattern at the beginning of all functions;
push %rbp
mov %rsp,%rbp
When frame pointers are available in a binary, full stack traces will be captured. Note that libraries that are frequently used by apps and ship with the operating system, such as libc, are generated in release mode and therefore do not include frame pointers. Frequently, when a backtrace includes an address from a system library, the backtrace will fail to resolve further as the frame pointer trail goes cold due to a missing frame pointer.
A simple application was developed to show the difference. The application calls function a(), which calls b(), which calls c(), etc. Function z() calls a heavy compute function called matrix_multiply(). Almost all of the IP samples are collected while matrix_multiple is executing. The next two screen shots show one of the main differences between frame pointers and LBRs.
Note that the frame pointer example, shows the full stack trace while the LBR example, only shows part of the stack due to the limited number of LBR registers in the CPU.
Kernel Samples
When an IP sample is captured while a kernel mode (i.e. operating system) function is executing, the sample will be shown with an address that starts with 0xffffffff and map to the [kernel.kallsyms] module.
[vdso]
Samples may be collected while a CPU is executing functions in the Virtual Dynamic Shared Object. In this case, the sample will be resolved (i.e. mapped) to the [vdso] module. The vdso man page provides the following description of the vdso:
The “vDSO“ (virtual dynamic shared object) is a small shared library
that the kernel automatically maps into the address space of all
user-space applications. Applications usually do not need to concern
themselves with these details as the vDSO is most commonly called by
the C library. This way you can code in the normal way using
standard functions and the C library will take care of using any
functionality that is available via the vDSO.
Why does the vDSO exist at all? There are some system calls the
kernel provides that user-space code ends up using frequently, to the
point that such calls can dominate overall performance. This is due
both to the frequency of the call as well as the context-switch
overhead that results from exiting user space and entering the
kernel.
[Unknown]
When an address can not be resolved (i.e. mapped to a module), its address within the process’ address space will be shown and its module will be marked as [Unknown].
Filter Dialog
Collapse unresolved lines is useful if some of the binary code does not have symbols. In this case, subtrees that consist of only unresolved symbols get collapsed in the Top-Down view, since they provide very little useful information.
Hide functions with CPU usage below X% is useful for large applications, where the sampling profiler hits lots of function just a few times. To filter out the “long tail,” which is typically not important for CPU performance bottleneck analysis, this checkbox should be selected.
Example of Using Timeline with Function Table
Here is an example walkthrough of using the timeline and function table with Instruction Pointer (IP)/backtrace Sampling Data
Timeline
When a collection result is opened in the Nsight Systems GUI, there are multiple ways to view the CPU profiling data - especially the CPU IP / backtrace data.
In the timeline, yellow-orange marks can be found under each thread’s timeline that indicate the moment an IP / backtrace sample was collected on that thread (e.g. see the yellow-orange marks in the Specific Samples box above). Hovering the cursor over a mark will cause a tooltip to display the backtrace for that sample.
Below the Timeline is a drop-down list with multiple options including Events View, Top-Down View, Bottom-Up View, and Flat View. All four of these views can be used to view CPU IP / backtrace sampling data.
If the Bottom-Up View is selected, here is the sampling summary shown in the bottom half of the Timeline View screen. Notice that the summary includes the phrase “65,022 samples are used” indicating how many samples are summarized. By default, functions that were found in less less than 0.5% of the samples are not show. Use the filter button to modify that setting.
When sampling data is filtered, the Sampling Summary will summarize the selected samples. Samples can be filtered on an OS thread basis, on a time basis, or both. Above, deselecting a checkbox next to a thread removes its samples from the sampling summary. Dragging the cursor over the timeline and selecting “Filter and Zoom In” chooses the samples during the time selected, as seen below. The sample summary includes the phrase “0.35% (225 samples) of data is shown due to applied filters” indicating that only 225 samples are included in the summary results.
Deselecting threads one at a time by deselecting their checkbox can be tedious. Click on the down arrow next to a thread and choose Show Only This Thread to deselect all threads except that thread.
If Events View is selected in the Timeline View’s drop-down list, right click on a specific thread and choose Show in Events View. The samples collected while that thread executed will be shown in the Events View. Double clicking on a specific sample in the Events view causes the timeline to show when that sample was collected - see the green boxes below. The backtrace for that sample is also shown in the Events View.
Backtraces
To understand the code path used to get to a specific function shown in the sampling summary, right click on a function and select Expand.
The above shows what happens when a function’s backtraces are expanded. In this case, the PCQueuePop function was called from the CmiGetNonLocal function which was called by the CsdNextMessage function which was called by the CsdScheduleForever function. The [Max depth] string marks the end of the collected backtrace.
Note that, by default, backtraces with less than 0.5% of the total backtraces are hidden. This behavior can make the percentage results hard to understand. If all backtraces are shown (i.e. the filter is disabled), the results look very different and the numbers add up as expected. To disable the filter, click on the Filter… button and uncheck the Hide functions with CPU usage below X% checkbox.
When the filter is disabled, the backtraces are recalculated. Note that you may need to right click on the function and select Expand again to get all of the backtraces to be shown.
When backtraces are collected, the whole sample (IP and backtrace) is handled as a single sample. If two samples have the exact same IP and backtrace, they are summed in the final results. If two samples have the same IP but a different backtrace, they will be shown as having the same leaf (i.e. IP) but a different backtrace. As mentioned earlier, when backtraces end, they are marked with the [Max depth] string (unless the backtrace can be traced back to its origin - e.g. __libc_start_main) or the backtrace breaks because an IP cannot be resolved.
Above, the leaf function is PCQueuePop. In this case, there are 11 different backtraces that lead to PCQueuPop - all of them end with [Max depth]. For example, the dominant path is PCQueuPop<-CmiGetNonLocal<-CsdNextmessage<-CsdScheduleForever<-[Max depth]. This path accounts for 5.67% of all samples as shown in line 5 (red numbers). The second most dominant path is PCQueuPop<-CmiGetNonLocal<-[Max depth] which accounts for 0.44% of all samples as shown in line 24 (red numbers). The path PCQueuPop<-CmiGetNonLocal<-CsdNextmessage<-CsdScheduleForever<-Sequencer::integrate(int)<-[Max depth] accounts for 0.03% of the samples as shown in line 7 (red numbers). Adding up percentages shown in the [Max depth] lines (lines 5, 7, 9, 13, 15, 16, 17, 19, 21, 23, and 24) generates 7.04% which equals the percentage of samples associated with the PCQueuePop function shown in line 0 (red numbers).
Multi-Report Timeline Views
Viewing Multiple Reports in Separate Panes
You have the option of looking at two or more Nsight Systems results files in separate panes. To do so, open each in a tab. Grab one of the tabs and undoc:
When you hover with that in the middle of the GUI, you will see options for where to dock the pane:
Multiple reports can be docked in the window.
Viewing Multiple Reports in the Same Timeline
You can open several reports in a single timeline. This could be done using one of these methods:
File > Open… in the main menu, and select several report files.
File > New multi-report view in the main menu, add report files that you want to open in the Multi-report view, and click the “Apply” button.
Multi-report view contains simple editor that allows to add/remove some report files and will load them all on a single timeline after applying that set of reports.
When reports are loaded, one can use the View Selector to open the Multi-report view again, change the set of reports, and click on “Apply” button to reload the timeline with the new set of reports.
The selected set of reports can be saved as a Multi-report view document and could be opened later to load the same set again.
Time Synchronization
When multiple reports are loaded into a single timeline, timestamps between them need to be adjusted, such that events that happened at the same time appear to be aligned.
Nsight Systems can automatically adjust timestamps based on UTC time
recorded around the collection start time. This method is used by default when
other more precise methods are not available. This time can be seen as UTC
time at t=0 in the Analysis Summary page of the report file. Refer to your
OS documentation to learn how to sync the software clock using the Network Time
Protocol (NTP). NTP-based time synchronization is not very precise, with the
typical errors on the scale of one to tens of milliseconds.
Reports collected on the same physical machine can use synchronization based on
Timestamp Counter (TSC) values. These are platform-specific counters,
typically accessed in user space applications using the RDTSC instruction on
x86_64 architecture, or by reading the CNTVCT register on Arm64. Their values
converted to nanoseconds can be seen as TSC value at t=0 in the Analysis
Summary page of the report file. Reports synchronized using TSC values can be
aligned with nanoseconds-level precision.
TSC-based time synchronization is activated automatically, when Nsight Systems
detects that reports come from same target and that the same TSC value
corresponds to very close UTC times. Targets are considered to be the same when
either explicitly set environment variables NSYS_HW_ID are the same for both
reports or when target hostnames are the same and NSYS_HW_ID is not set for
either target. The difference between UTC and TSC time offsets must be below 1
second to choose TSC-based time synchronization.
To find out which synchronization method was used, navigate to the Analysis
Summary tab of an added report and check the Report alignment source
property of a target. Note, that the first report won’t have this parameter.
When loading multiple reports into a single timeline, it is always advisable to first check that time synchronization looks correct, by zooming into synchronization or communication events that are expected to be aligned.
Timeline Hierarchy
When reports are added to the same timeline Nsight Systems will automatically
line them up by timestamps as described above. If you want Nsight Systems to
also recognize matching process or hardware information, you will need to set
environment variables NSYS_SYSTEM_ID and NSYS_HW_ID as shown below at
the time of report collection (such as when using “nsys profile …” command).
When loading a pair of given report files into the same timeline, they will be merged in one of the following configurations:
Different hardware — is used when reports are coming from different physical machines, and no hardware resources are shared in these reports. This mode is used when neither
NSYS_HW_IDorNSYS_SYSTEM_IDis set and target hostnames are different or absent, and can be additionally signalled by specifying differentNSYS_HW_IDvalues.Different systems, same hardware — is used when reports are collected on different virtual machines (VMs) or containers on the same physical machine. To activate this mode, specify the same value of
NSYS_HW_IDwhen collecting the reports.Same system — is used when reports are collected within the same operating system (or container) environment. In this mode a process identifier (PID) 100 will refer to the same process in both reports. To manually activate this mode, specify the same value of
NSYS_SYSTEM_IDwhen collecting the reports. This mode is automatically selected when target hostnames are the same and neitherNSYS_HW_IDorNSYS_SYSTEM_IDis provided.
The following diagrams demonstrate typical cases:
Example: MPI
A typical scenario is when a computing job is run using one of the MPI implementations. Each instance of the app can be profiled separately, resulting in multiple report files. For example:
# Run MPI job without the profiler:
mpirun <mpirun-options> ./myApp
# Run MPI job and profile each instance of the application:
mpirun <mpirun-options> nsys profile -o report-%p <nsys-options>./myApp
When each MPI rank runs on a different node, the command above works fine, since the default pairing mode (different hardware) will be used.
When all MPI ranks run the localhost only, use this command (value “A” was chosen arbitrarily, it can be any non-empty string):
NSYS_SYSTEM_ID=A mpirun <mpirun-options> nsys profile -o report-%p < nsys -options> ./myApp
For convenience, the MPI rank can be encoded into the report filename. For Open MPI, use the following command to create report files based on the global rank value:
mpirun <mpirun-options> nsys profile -o report-%q{OMPI_COMM_WORLD_RANK} < nsys -options> ./myApp
MPICH-based implementations set the environment variable PMI_RANK and Slurm
(srun) provides the global MPI rank in SLURM_PROCID.
Limitations on Syncing Multiple Reports in Timeline
Only report files collected with Nsight Systems version 2021.3 and newer are fully supported.
Sequential reports collected in a single CLI profiling session cannot be loaded into a single timeline yet.
Add-on Graphs - Flame Graph
The generation of Flame Graphs from Nsight Systems reports is not a built-in feature, but it is possible to create such graphs from Nsight Systems reports with the script stackcollapse_nsys.py located at <nsys-install-dir>/<host-folder>/Scripts/Flamegraph. There is also a README.md file at that location with additional usage details.
Requirements:
flamegraph.plfrom Brendan Gregg’s FlameGraph github,Perl
Usage
Generating flamegraph from Nsight Systems report file:
Linux
`
python3 stackcollapse_nsys.py report.nsys-rep | ./flamegraph.pl > result_flamegraph.svg
`
Windows
`
PowerShell -Command "python stackcollapse_nsys.py report.nsys-rep | perl flamegraph.pl > result_flamegraph.svg"
`
The script exports the report to SQLite, queries the CPU samples and passes them as input to flamegraph.pl.
Parameters
The following parameters can be passed to the script:
Short |
Long |
Default |
Switch Description |
|---|---|---|---|
|
Current Nsight Systems CLI installation location |
Path to the Nsight Systems CLI directory (for ex: /opt/nvidia/nsight-systems/2022.4.1/target-linux-x64). |
|
-o |
|
Output is written to stdout |
Path to a result file containing a data suitable for flamegraph.pl. |
|
False |
Use full function names with return type, arguments and expanded templates, if available. |
Note
By default, the script tries to shorten function definitions (eliminating
return type, arguments and templates). In some complex cases
shortening may fail and return a full function definition. To disable
shortening defining --full_function_names=False argument can be used.
Here is an example of a Flame Graph generated from an Nsight Systems report. The program was a debug build of GROMACS, running on two ranks, each running two OpenMP threads.
Post-Collection Analysis
Once you have profiled using Nsight Systems there are many options for analyzing the collected data as well as to output it in various formats. These options are available from the CLI or the GUI
Available Export Formats
SQLite Schema Reference
Nsight Systems has the ability to export SQLite database files from the .nsys-rep
results file. From the CLI, use nsys export. From the GUI, call
File->Export....
Note: The .nsys-rep report format is the only data format for Nsight Systems that should be considered forward compatible. The SQLite schema can and will change in the future.
The schema for a concrete database can be obtained with the sqlite3 tool
built-in command .schema. The sqlite3 tool can be located in the Target or
Host directory of your Nsight Systems installation.
Note: Currently tables are created lazily, and therefore not every table
described in the documentation will be present in a particular database. This
will change in a future version of the product. If you want a full schema of all
possible tables, use nsys export --lazy=false during export phase.
Currently, a table is created for each data type in the exported database. Since usage patterns for exported data may vary greatly and no default use cases have been established, no indexes or extra constraints are created. Instead, refer to the SQLite Examples section for a list of common recipes. This may change in a future version of the product.
To check the version of your exported SQLite file, check the value of
EXPORT_SCHEMA_VERSION in the META_DATA_EXPORT table. The schema version
is a common three-value major/minor/micro version number. The first value, or
major value, indicates the overall format of the database, and is only changed
if there is a major re-write or re-factor of the entire database format. It is
assumed that if the major version changes, all scripts or queries will break.
The middle, or minor, version is changed anytime there is a more localized, but
potentially breaking change, such as renaming an existing column, or changing
the type of an existing column. The last, or micro version is changed any time
there are additions, such as a new table or column, that should not introduce
any breaking change when used with well-written, best-practices queries.
This is the schema as of the 2024.2 release, schema version 3.11.0.
CREATE TABLE StringIds (
-- Consolidation of repetitive string values.
id INTEGER NOT NULL PRIMARY KEY, -- ID reference value.
value TEXT NOT NULL -- String value.
);
CREATE TABLE ANALYSIS_FILE (
-- Analysis file content
id INTEGER NOT NULL PRIMARY KEY, -- ID reference value.
filename TEXT NOT NULL, -- File path
content TEXT NOT NULL, -- File content
globalPid INTEGER NOT NULL -- Serialized GlobalId.
);
CREATE TABLE ThreadNames (
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Thread name
priority INTEGER, -- Priority of the thread.
globalTid INTEGER -- Serialized GlobalId.
);
CREATE TABLE ProcessStreams (
globalPid INTEGER NOT NULL, -- Serialized GlobalId.
filenameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- File name
contentId INTEGER NOT NULL -- REFERENCES StringIds(id) -- Stream content
);
CREATE TABLE TARGET_INFO_SYSTEM_ENV (
globalVid INTEGER, -- Serialized GlobalId.
devStateName TEXT NOT NULL, -- Device state name.
name TEXT NOT NULL, -- Property name.
nameEnum INTEGER NOT NULL, -- Property enum value.
value TEXT NOT NULL -- Property value.
);
CREATE TABLE TARGET_INFO_NIC_INFO (
globalId INTEGER, -- Device state globalId
stateName TEXT NOT NULL, -- Device state name
nicId INTEGER NOT NULL, -- Network interface Id.
name TEXT NOT NULL, -- Network interface name
deviceId INTEGER NOT NULL, -- REFERENCES ENUM_NET_DEVICE_ID(id)
vendorId INTEGER NOT NULL, -- REFERENCES ENUM_NET_VENDOR_ID(id)
linkLayer INTEGER NOT NULL -- REFERENCES ENUM_NET_LINK_TYPE(id)
);
CREATE TABLE NIC_ID_MAP (
-- Map between NIC info nicId and NIC metric globalId
nicId INTEGER NOT NULL, -- REFERENCES TARGET_INFO_NIC_INFO(nicId)
globalId INTEGER NOT NULL -- REFERENCES NET_NIC_METRIC(globalId)
);
CREATE TABLE TARGET_INFO_SESSION_START_TIME (
utcEpochNs INTEGER, -- UTC Epoch timestamp at start of the capture (ns).
utcTime TEXT, -- Start of the capture in UTC.
localTime TEXT -- Start of the capture in local time of target.
);
CREATE TABLE ANALYSIS_DETAILS (
-- Details about the analysis session.
globalVid INTEGER NOT NULL, -- Serialized GlobalId.
duration INTEGER NOT NULL, -- The total time span of the entire trace (ns).
startTime INTEGER NOT NULL, -- Trace start timestamp in nanoseconds.
stopTime INTEGER NOT NULL -- Trace stop timestamp in nanoseconds.
);
CREATE TABLE PMU_EVENT_REQUESTS (
-- PMU event requests
id INTEGER NOT NULL, -- PMU event request.
eventid INTEGER, -- PMU counter event id.
source INTEGER NOT NULL, -- REFERENCES ENUM_PMU_EVENT_SOURCE(id)
unit_type INTEGER NOT NULL, -- REFERENCES ENUM_PMU_UNIT_TYPE(id)
event_name TEXT, -- PMU counter unique name
PRIMARY KEY (id)
);
CREATE TABLE TARGET_INFO_GPU (
vmId INTEGER NOT NULL, -- Serialized GlobalId.
id INTEGER NOT NULL, -- Device ID.
name TEXT, -- Device name.
busLocation TEXT, -- PCI bus location.
isDiscrete INTEGER, -- True if discrete, false if integrated.
l2CacheSize INTEGER, -- Size of L2 cache (B).
totalMemory INTEGER, -- Total amount of memory on the device (B).
memoryBandwidth INTEGER, -- Amount of memory transferred (B).
clockRate INTEGER, -- Clock frequency (Hz).
smCount INTEGER, -- Number of multiprocessors on the device.
pwGpuId INTEGER, -- PerfWorks GPU ID.
uuid TEXT, -- Device UUID.
luid INTEGER, -- Device LUID.
chipName TEXT, -- Chip name.
cuDevice INTEGER, -- CUDA device ID.
ctxswDevPath TEXT, -- GPU context switch device node path.
ctrlDevPath TEXT, -- GPU control device node path.
revision INTEGER, -- Revision number.
nodeMask INTEGER, -- Device node mask.
constantMemory INTEGER, -- Memory available on device for __constant__ variables (B).
maxIPC INTEGER, -- Maximum instructions per count.
maxRegistersPerBlock INTEGER, -- Maximum number of 32-bit registers available per block.
maxShmemPerBlock INTEGER, -- Maximum optin shared memory per block.
maxShmemPerBlockOptin INTEGER, -- Maximum optin shared memory per block.
maxShmemPerSm INTEGER, -- Maximum shared memory available per multiprocessor (B).
maxRegistersPerSm INTEGER, -- Maximum number of 32-bit registers available per multiprocessor.
threadsPerWarp INTEGER, -- Warp size in threads.
asyncEngines INTEGER, -- Number of asynchronous engines.
maxWarpsPerSm INTEGER, -- Maximum number of warps per multiprocessor.
maxBlocksPerSm INTEGER, -- Maximum number of blocks per multiprocessor.
maxThreadsPerBlock INTEGER, -- Maximum number of threads per block.
maxBlockDimX INTEGER, -- Maximum X-dimension of a block.
maxBlockDimY INTEGER, -- Maximum Y-dimension of a block.
maxBlockDimZ INTEGER, -- Maximum Z-dimension of a block.
maxGridDimX INTEGER, -- Maximum X-dimension of a grid.
maxGridDimY INTEGER, -- Maximum Y-dimension of a grid.
maxGridDimZ INTEGER, -- Maximum Z-dimension of a grid.
computeMajor INTEGER, -- Major compute capability version number.
computeMinor INTEGER, -- Minor compute capability version number.
smMajor INTEGER, -- Major multiprocessor version number.
smMinor INTEGER -- Minor multiprocessor version number.
);
CREATE TABLE TARGET_INFO_XMC_SPEC (
vmId INTEGER NOT NULL, -- Serialized GlobalId.
clientId INTEGER NOT NULL, -- Client ID.
type TEXT NOT NULL, -- Client type.
name TEXT NOT NULL, -- Client name.
groupId TEXT NOT NULL -- Client group ID.
);
CREATE TABLE TARGET_INFO_PROCESS (
processId INTEGER NOT NULL, -- Process ID.
openGlVersion TEXT NOT NULL, -- OpenGL version.
correlationId INTEGER NOT NULL, -- Correlation ID of the kernel.
nameId INTEGER NOT NULL -- REFERENCES StringIds(id) -- Function name
);
CREATE TABLE TARGET_INFO_NVTX_CUDA_DEVICE (
name TEXT NOT NULL, -- CUDA device name assigned using NVTX.
hwId INTEGER NOT NULL, -- Hardware ID.
vmId INTEGER NOT NULL, -- VM ID.
deviceId INTEGER NOT NULL -- Device ID.
);
CREATE TABLE TARGET_INFO_NVTX_CUDA_CONTEXT (
name TEXT NOT NULL, -- CUDA context name assigned using NVTX.
hwId INTEGER NOT NULL, -- Hardware ID.
vmId INTEGER NOT NULL, -- VM ID.
processId INTEGER NOT NULL, -- Process ID.
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL -- Context ID.
);
CREATE TABLE TARGET_INFO_NVTX_CUDA_STREAM (
name TEXT NOT NULL, -- CUDA stream name assigned using NVTX.
hwId INTEGER NOT NULL, -- Hardware ID.
vmId INTEGER NOT NULL, -- VM ID.
processId INTEGER NOT NULL, -- Process ID.
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
streamId INTEGER NOT NULL -- Stream ID.
);
CREATE TABLE TARGET_INFO_CUDA_CONTEXT_INFO (
nullStreamId INTEGER NOT NULL, -- Stream ID.
hwId INTEGER NOT NULL, -- Hardware ID.
vmId INTEGER NOT NULL, -- VM ID.
processId INTEGER NOT NULL, -- Process ID.
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
parentContextId INTEGER, -- For green context, this is the parent context id.
isGreenContext INTEGER -- Is this a Green Context?
);
CREATE TABLE TARGET_INFO_CUDA_STREAM (
streamId INTEGER NOT NULL, -- Stream ID.
hwId INTEGER NOT NULL, -- Hardware ID.
vmId INTEGER NOT NULL, -- VM ID.
processId INTEGER NOT NULL, -- Process ID.
contextId INTEGER NOT NULL, -- Context ID.
priority INTEGER NOT NULL, -- Priority of the stream.
flag INTEGER NOT NULL -- REFERENCES ENUM_CUPTI_STREAM_TYPE(id)
);
CREATE TABLE TARGET_INFO_WDDM_CONTEXTS (
context INTEGER NOT NULL,
engineType INTEGER NOT NULL,
nodeOrdinal INTEGER NOT NULL,
friendlyName TEXT NOT NULL
);
CREATE TABLE TARGET_INFO_PERF_METRIC (
id INTEGER NOT NULL, -- Event or Metric ID value
name TEXT NOT NULL, -- Event or Metric name
description TEXT NOT NULL, -- Event or Metric description
unit TEXT NOT NULL -- Event or Metric measurement unit
);
CREATE TABLE TARGET_INFO_COMPONENT (
componentId INTEGER NOT NULL, -- Component ID value
name TEXT NOT NULL, -- Component name
instance INTEGER NOT NULL, -- Component instance
parentComponentId INTEGER NOT NULL -- Parent Component ID value
);
CREATE TABLE TARGET_INFO_NETWORK_METRICS (
metricsListId INTEGER NOT NULL, -- Metric list ID
metricsIdx INTEGER NOT NULL, -- List index of metric
name TEXT NOT NULL, -- Name of metric
description TEXT NOT NULL, -- Description of metric
unit TEXT NOT NULL -- Measurement unit of metric
);
CREATE TABLE META_DATA_CAPTURE (
-- information about nsys capture parameters
name TEXT NOT NULL, -- Name of meta-data record
value TEXT -- Value of meta-data record
);
CREATE TABLE META_DATA_EXPORT (
-- information about nsys export process
name TEXT NOT NULL, -- Name of meta-data record
value TEXT -- Value of meta-data record
);
CREATE TABLE ENUM_NSYS_EVENT_TYPE (
-- Nsys event type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_NSYS_EVENT_CLASS (
-- Nsys event class labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_GPU_CTX_SWITCH (
-- GPU context switch labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_MEMCPY_OPER (
-- CUDA memcpy operation labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_MEM_KIND (
-- CUDA memory kind labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_MEMPOOL_TYPE (
-- CUDA mempool type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_MEMPOOL_OPER (
-- CUDA mempool operation labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_DEV_MEM_EVENT_OPER (
-- CUDA device mem event operation labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_KERNEL_LAUNCH_TYPE (
-- CUDA kernel launch type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_SHARED_MEM_LIMIT_CONFIG (
-- CUDA shared memory limit config labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_UNIF_MEM_MIGRATION (
-- CUDA unified memory migration cause labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_UNIF_MEM_ACCESS_TYPE (
-- CUDA unified memory access type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUDA_FUNC_CACHE_CONFIG (
-- CUDA function cache config labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUPTI_STREAM_TYPE (
-- CUPTI stream type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_CUPTI_SYNC_TYPE (
-- CUPTI synchronization type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_STACK_UNWIND_METHOD (
-- Stack unwind method labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_SAMPLING_THREAD_STATE (
-- Sampling thread state labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENGL_DEBUG_SOURCE (
-- OpenGL debug source labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENGL_DEBUG_TYPE (
-- OpenGL debug type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENGL_DEBUG_SEVERITY (
-- OpenGL debug severity labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_VULKAN_PIPELINE_CREATION_FLAGS (
-- Vulkan pipeline creation feedback flag labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_D3D12_HEAP_TYPE (
-- D3D12 heap type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_D3D12_PAGE_PROPERTY (
-- D3D12 CPU page property labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_D3D12_HEAP_FLAGS (
-- D3D12 heap flag labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_D3D12_CMD_LIST_TYPE (
-- D3D12 command list type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENACC_DEVICE (
-- OpenACC device type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENACC_EVENT_KIND (
-- OpenACC event type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_EVENT_KIND (
-- OpenMP event kind labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_THREAD (
-- OpenMP thread labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_DISPATCH (
-- OpenMP dispatch labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_SYNC_REGION (
-- OpenMP sync region labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_WORK (
-- OpenMP work labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_MUTEX (
-- OpenMP mutex labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_TASK_FLAG (
-- OpenMP task flags labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_OPENMP_TASK_STATUS (
-- OpenMP task status labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_SLI_TRANSER (
-- OpenMP task status labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_DXGI_FORMAT (
-- DXGI image format labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_NVDRIVER_EVENT_ID (
-- NV-Driver event it labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_WDDM_PAGING_QUEUE_TYPE (
-- WDDM paging queue type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_WDDM_PACKET_TYPE (
-- WDDM packet type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_WDDM_ENGINE_TYPE (
-- WDDM engine type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_WDDM_INTERRUPT_TYPE (
-- WDDM DMA interrupt type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_WDDM_VIDMM_OP_TYPE (
-- WDDM VidMm operation type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_NET_LINK_TYPE (
-- NIC link layer labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_NET_DEVICE_ID (
-- NIC PCIe device id labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_NET_VENDOR_ID (
-- NIC PCIe vendor id labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_ETW_MEMORY_TRANSFER_TYPE (
-- memory transfer type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value
name TEXT, -- Enum symbol name
label TEXT -- Enum human name
);
CREATE TABLE ENUM_PMU_EVENT_SOURCE (
-- PMU event source labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE ENUM_PMU_UNIT_TYPE (
-- PMU unit type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE ENUM_VIDEO_ENGINE_TYPE (
-- Video engine type id labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE ENUM_VIDEO_ENGINE_CODEC (
-- Video engine codec labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE ENUM_DIAGNOSTIC_SEVERITY_LEVEL (
-- Diagnostic message severity level labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE ENUM_DIAGNOSTIC_SOURCE_TYPE (
-- Diagnostic message source type labels
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE ENUM_DIAGNOSTIC_TIMESTAMP_SOURCE (
-- Diagnostic message timestamp source lables
id INTEGER NOT NULL PRIMARY KEY, -- Enum numerical value.
name TEXT, -- Enum symbol name.
label TEXT -- Enum human name.
);
CREATE TABLE GENERIC_EVENT_SOURCES (
sourceId INTEGER NOT NULL PRIMARY KEY, -- Serialized GlobalId.
data TEXT NOT NULL -- JSON encoded generic event source description.
);
CREATE TABLE GENERIC_EVENT_TYPES (
typeId INTEGER NOT NULL PRIMARY KEY, -- Serialized GlobalId.
sourceId INTEGER NOT NULL, -- REFERENCES GENERIC_EVENT_SOURCES(sourceId)
data TEXT NOT NULL -- JSON encoded generic event type description.
);
CREATE TABLE GENERIC_EVENTS (
-- Events for which the schema is known only at the run-time.
rawTimestamp INTEGER NOT NULL, -- Raw event timestamp recorded during profiling.
timestamp INTEGER, -- Event timestamp converted to the profiling session timeline.
typeId INTEGER NOT NULL, -- REFERENCES GENERIC_EVENT_TYPES(typeId)
data TEXT NOT NULL -- JSON encoded event data.
);
CREATE TABLE ETW_PROVIDERS (
-- Names and identifiers of ETW providers captured in the report.
providerId INTEGER NOT NULL PRIMARY KEY, -- Provider ID.
providerNameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Provider name
guid TEXT NOT NULL -- ETW Provider GUID.
);
CREATE TABLE ETW_TASKS (
-- Names and identifiers of ETW tasks captured in the report.
taskNameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Task name
taskId INTEGER NOT NULL, -- The event task ID.
providerId INTEGER NOT NULL -- Provider ID.
);
CREATE TABLE ETW_EVENTS (
-- Raw ETW events captured in the report.
timestamp INTEGER NOT NULL, -- Event start timestamp (ns).
processId INTEGER, -- Process ID.
threadId INTEGER, -- Thread ID.
providerId INTEGER NOT NULL, -- Provider ID.
taskId INTEGER NOT NULL, -- The event task ID.
eventId INTEGER NOT NULL, -- Event ID.
version INTEGER NOT NULL, -- The event version.
opcode INTEGER, -- The event opcode.
data TEXT NOT NULL -- JSON encoded event data.
);
CREATE TABLE TARGET_INFO_GPU_METRICS (
-- GPU Metrics, metric names and ids.
typeId INTEGER NOT NULL, -- REFERENCES GENERIC_EVENT_TYPES(typeId)
sourceId INTEGER NOT NULL, -- REFERENCES GENERIC_EVENT_SOURCES(sourceId)
typeName TEXT NOT NULL, -- Name of event type.
metricId INTEGER NOT NULL, -- Id of metric in event; not assumed to be stable.
metricName TEXT NOT NULL -- Definitive name of metric.
);
CREATE TABLE GPU_METRICS (
-- GPU Metrics, events and values.
timestamp INTEGER, -- Event timestamp (ns).
typeId INTEGER NOT NULL, -- REFERENCES TARGET_INFO_GPU_METRICS(typeId) and GENERIC_EVENT_TYPES(typeId)
metricId INTEGER NOT NULL, -- REFERENCES TARGET_INFO_GPU_METRICS(metricId)
value INTEGER NOT NULL -- Counter data value
);
CREATE TABLE MPI_COMMUNICATORS (
-- Identification of MPI communication groups.
rank INTEGER, -- Active MPI rank
timestamp INTEGER, -- Time of MPI communicator creation.
commHandle INTEGER, -- MPI communicator handle
parentHandle INTEGER, -- MPI communicator handle
localRank INTEGER, -- Local MPI rank in a communicator.
size INTEGER, -- MPI communicator size.
groupRoot INTEGER, -- Root rank (global) in MPI communicator.
groupRootUid INTEGER, -- Group root's communicator ID.
members TEXT -- MPI communicator members (index is global rank).
);
CREATE TABLE CUPTI_ACTIVITY_KIND_MEMCPY (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
streamId INTEGER NOT NULL, -- Stream ID.
correlationId INTEGER, -- REFERENCES CUPTI_ACTIVITY_KIND_RUNTIME(correlationId)
globalPid INTEGER, -- Serialized GlobalId.
bytes INTEGER NOT NULL, -- Number of bytes transferred (B).
copyKind INTEGER NOT NULL, -- REFERENCES ENUM_CUDA_MEMCPY_OPER(id)
deprecatedSrcId INTEGER, -- Deprecated, use srcDeviceId instead.
srcKind INTEGER, -- REFERENCES ENUM_CUDA_MEM_KIND(id)
dstKind INTEGER, -- REFERENCES ENUM_CUDA_MEM_KIND(id)
srcDeviceId INTEGER, -- Source device ID.
srcContextId INTEGER, -- Source context ID.
dstDeviceId INTEGER, -- Destination device ID.
dstContextId INTEGER, -- Destination context ID.
migrationCause INTEGER, -- REFERENCES ENUM_CUDA_UNIF_MEM_MIGRATION(id)
graphNodeId INTEGER, -- REFERENCES CUDA_GRAPH_NODE_EVENTS(graphNodeId)
virtualAddress INTEGER -- Virtual base address of the page/s being transferred.
);
CREATE TABLE CUPTI_ACTIVITY_KIND_MEMSET (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
streamId INTEGER NOT NULL, -- Stream ID.
correlationId INTEGER, -- REFERENCES CUPTI_ACTIVITY_KIND_RUNTIME(correlationId)
globalPid INTEGER, -- Serialized GlobalId.
value INTEGER NOT NULL, -- Value assigned to memory.
bytes INTEGER NOT NULL, -- Number of bytes set (B).
graphNodeId INTEGER, -- REFERENCES CUDA_GRAPH_NODE_EVENTS(graphNodeId)
memKind INTEGER -- REFERENCES ENUM_CUDA_MEM_KIND(id)
);
CREATE TABLE CUPTI_ACTIVITY_KIND_KERNEL (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
streamId INTEGER NOT NULL, -- Stream ID.
correlationId INTEGER, -- REFERENCES CUPTI_ACTIVITY_KIND_RUNTIME(correlationId)
globalPid INTEGER, -- Serialized GlobalId.
demangledName INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Kernel function name w/ templates
shortName INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Base kernel function name
mangledName INTEGER, -- REFERENCES StringIds(id) -- Raw C++ mangled kernel function name
launchType INTEGER, -- REFERENCES ENUM_CUDA_KERNEL_LAUNCH_TYPE(id)
cacheConfig INTEGER, -- REFERENCES ENUM_CUDA_FUNC_CACHE_CONFIG(id)
registersPerThread INTEGER NOT NULL, -- Number of registers required for each thread executing the kernel.
gridX INTEGER NOT NULL, -- X-dimension grid size.
gridY INTEGER NOT NULL, -- Y-dimension grid size.
gridZ INTEGER NOT NULL, -- Z-dimension grid size.
blockX INTEGER NOT NULL, -- X-dimension block size.
blockY INTEGER NOT NULL, -- Y-dimension block size.
blockZ INTEGER NOT NULL, -- Z-dimension block size.
staticSharedMemory INTEGER NOT NULL, -- Static shared memory allocated for the kernel (B).
dynamicSharedMemory INTEGER NOT NULL, -- Dynamic shared memory reserved for the kernel (B).
localMemoryPerThread INTEGER NOT NULL, -- Amount of local memory reserved for each thread (B).
localMemoryTotal INTEGER NOT NULL, -- Total amount of local memory reserved for the kernel (B).
gridId INTEGER NOT NULL, -- Unique grid ID of the kernel assigned at runtime.
sharedMemoryExecuted INTEGER, -- Shared memory size set by the driver.
graphNodeId INTEGER, -- REFERENCES CUDA_GRAPH_NODE_EVENTS(graphNodeId)
sharedMemoryLimitConfig INTEGER -- REFERENCES ENUM_CUDA_SHARED_MEM_LIMIT_CONFIG(id)
);
CREATE TABLE CUPTI_ACTIVITY_KIND_SYNCHRONIZATION (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
streamId INTEGER NOT NULL, -- Stream ID.
correlationId INTEGER, -- Correlation ID of the synchronization API to which this result is associated.
globalPid INTEGER, -- Serialized GlobalId.
syncType INTEGER NOT NULL, -- REFERENCES ENUM_CUPTI_SYNC_TYPE(id)
eventId INTEGER NOT NULL -- Event ID for which the synchronization API is called.
);
CREATE TABLE CUPTI_ACTIVITY_KIND_CUDA_EVENT (
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
streamId INTEGER NOT NULL, -- Stream ID.
correlationId INTEGER, -- Correlation ID of the event record API to which this result is associated.
globalPid INTEGER, -- Serialized GlobalId.
eventId INTEGER NOT NULL -- Event ID for which the event record API is called.
);
CREATE TABLE CUPTI_ACTIVITY_KIND_GRAPH_TRACE (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
greenContextId INTEGER, -- Green context ID.
streamId INTEGER NOT NULL, -- Stream ID.
correlationId INTEGER, -- REFERENCES CUPTI_ACTIVITY_KIND_RUNTIME(correlationId)
globalPid INTEGER, -- Serialized GlobalId.
graphId INTEGER NOT NULL, -- REFERENCES CUDA_GRAPH_EVENTS(graphId)
graphExecId INTEGER NOT NULL -- REFERENCES CUDA_GRAPH_EVENTS(graphExecId)
);
CREATE TABLE CUPTI_ACTIVITY_KIND_RUNTIME (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- ID used to identify events that this function call has triggered.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
returnValue INTEGER NOT NULL, -- Return value of the function call.
callchainId INTEGER -- REFERENCES CUDA_CALLCHAINS(id)
);
CREATE TABLE CUDNN_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
nameId INTEGER NOT NULL -- REFERENCES StringIds(id) -- Function name
);
CREATE TABLE CUBLAS_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
nameId INTEGER NOT NULL -- REFERENCES StringIds(id) -- Function name
);
CREATE TABLE CUDA_GRAPH_NODE_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
graphNodeId INTEGER NOT NULL, -- REFERENCES CUDA_GRAPH_NODE_EVENTS(graphNodeId)
originalGraphNodeId INTEGER -- Reference to the original graph node ID, if cloned node.
);
CREATE TABLE CUDA_GRAPH_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
graphId INTEGER, -- Graph ID.
originalGraphId INTEGER, -- Reference to the original graph ID, if cloned.
graphExecId INTEGER -- Executable graph ID.
);
CREATE TABLE CUDA_UM_CPU_PAGE_FAULT_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
globalPid INTEGER NOT NULL, -- Serialized GlobalId.
address INTEGER NOT NULL, -- Virtual address of the page that faulted.
originalFaultPc INTEGER, -- Program counter of the CPU instruction that caused the page fault.
CpuInstruction INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
module INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Module name
unresolvedFaultPc INTEGER -- True if the program counter was not resolved.
);
CREATE TABLE CUDA_UM_GPU_PAGE_FAULT_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
globalPid INTEGER NOT NULL, -- Serialized GlobalId.
deviceId INTEGER NOT NULL, -- Device ID.
address INTEGER NOT NULL, -- Virtual address of the page that faulted.
numberOfPageFaults INTEGER NOT NULL, -- Number of page faults for the same page.
faultAccessType INTEGER NOT NULL -- REFERENCES ENUM_CUDA_UNIF_MEM_ACCESS_TYPE(id)
);
CREATE TABLE CUDA_GPU_MEMORY_USAGE_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
globalPid INTEGER NOT NULL, -- Serialized GlobalId.
deviceId INTEGER NOT NULL, -- Device ID.
contextId INTEGER NOT NULL, -- Context ID.
address INTEGER NOT NULL, -- Virtual address of the allocation/deallocation.
pc INTEGER NOT NULL, -- Program counter of the allocation/deallocation.
bytes INTEGER NOT NULL, -- Number of bytes allocated/deallocated (B).
memKind INTEGER NOT NULL, -- REFERENCES ENUM_CUDA_MEM_KIND(id)
memoryOperationType INTEGER NOT NULL, -- REFERENCES ENUM_CUDA_DEV_MEM_EVENT_OPER(id)
name TEXT, -- Variable name, if available.
correlationId INTEGER, -- REFERENCES CUPTI_ACTIVITY_KIND_RUNTIME(correlationId)
localMemoryPoolAddress INTEGER, -- Base address of the local memory pool used
localMemoryPoolReleaseThreshold INTEGER, -- Release threshold of the local memory pool used
localMemoryPoolSize INTEGER, -- Size of the local memory pool used
localMemoryPoolUtilizedSize INTEGER, -- Utilized size of the local memory pool used
importedMemoryPoolAddress INTEGER, -- Base address of the imported memory pool used
importedMemoryPoolProcessId INTEGER -- Process ID of the imported memory pool used
);
CREATE TABLE CUDA_GPU_MEMORY_POOL_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
globalPid INTEGER NOT NULL, -- Serialized GlobalId.
deviceId INTEGER NOT NULL, -- Device ID.
address INTEGER NOT NULL, -- The base virtual address of the memory pool.
operationType INTEGER NOT NULL, -- REFERENCES ENUM_CUDA_MEMPOOL_OPER(id)
poolType INTEGER NOT NULL, -- REFERENCES ENUM_CUDA_MEMPOOL_TYPE(id)
correlationId INTEGER, -- REFERENCES CUPTI_ACTIVITY_KIND_RUNTIME(correlationId)
minBytesToKeep INTEGER, -- Minimum number of bytes to keep of the memory pool.
localMemoryPoolReleaseThreshold INTEGER, -- Release threshold of the local memory pool used
localMemoryPoolSize INTEGER, -- Size of the local memory pool used
localMemoryPoolUtilizedSize INTEGER -- Utilized size of the local memory pool used
);
CREATE TABLE CUDA_CALLCHAINS (
id INTEGER NOT NULL, -- Part of PRIMARY KEY (id, stackDepth).
symbol INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
module INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Module name
unresolved INTEGER, -- True if the symbol was not resolved.
originalIP INTEGER, -- Instruction pointer value.
stackDepth INTEGER NOT NULL, -- Zero-base index of the given function in call stack.
PRIMARY KEY (id, stackDepth)
);
CREATE TABLE MPI_RANKS (
-- Mapping of global thread IDs (gtid) to MPI ranks
globalTid INTEGER NOT NULL, -- Serialized GlobalId.
rank INTEGER NOT NULL -- MPI rank
);
CREATE TABLE MPI_P2P_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER, -- REFERENCES StringIds(id) -- Registered NVTX domain
commHandle INTEGER, -- MPI communicator handle
tag INTEGER, -- MPI message tag
remoteRank INTEGER, -- MPI remote rank (destination or source)
size INTEGER, -- MPI message size in bytes
requestHandle INTEGER -- MPI request handle
);
CREATE TABLE MPI_COLLECTIVES_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER, -- REFERENCES StringIds(id) -- Registered NVTX domain
commHandle INTEGER, -- MPI communicator handle
rootRank INTEGER, -- root rank in the collective
size INTEGER, -- MPI message size in bytes (send size for bidirectional ops)
recvSize INTEGER, -- MPI receive size in bytes
requestHandle INTEGER -- MPI request handle
);
CREATE TABLE MPI_START_WAIT_EVENTS (
-- MPI_Wait* and MPI_Start*
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER, -- REFERENCES StringIds(id) -- Registered NVTX domain
requestHandle INTEGER -- MPI request handle
);
CREATE TABLE MPI_OTHER_EVENTS (
-- MPI events without additional parameters
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER -- REFERENCES StringIds(id) -- Registered NVTX domain
);
CREATE TABLE UCP_WORKERS (
globalTid INTEGER NOT NULL, -- Serialized GlobalId.
workerUid INTEGER NOT NULL -- UCP worker UID
);
CREATE TABLE UCP_SUBMIT_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER, -- REFERENCES StringIds(id) -- Registered NVTX domain
bufferAddr INTEGER, -- Address of the message buffer
packedSize INTEGER, -- Message size (packed) in bytes
peerWorkerUid INTEGER, -- Peer's UCP worker UID
tag INTEGER -- UCP message tag
);
CREATE TABLE UCP_PROGRESS_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER, -- REFERENCES StringIds(id) -- Registered NVTX domain
bufferAddr INTEGER, -- Address of the message buffer
packedSize INTEGER, -- Message size (packed) in bytes
peerWorkerUid INTEGER, -- Peer's UCP worker UID
tag INTEGER -- UCP message tag
);
CREATE TABLE UCP_EVENTS (
-- UCP events without additional parameters
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
textId INTEGER -- REFERENCES StringIds(id) -- Registered NVTX domain
);
CREATE TABLE NVTX_EVENTS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
eventType INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_TYPE(id)
rangeId INTEGER, -- Correlation ID returned from a nvtxRangeStart call.
category INTEGER, -- User-controlled ID that can be used to group events.
color INTEGER, -- Encoded ARGB color value.
text TEXT, -- Optional text message for non registered strings.
globalTid INTEGER, -- Serialized GlobalId.
endGlobalTid INTEGER, -- Serialized GlobalId.
textId INTEGER, -- REFERENCES StringIds(id) -- Registered NVTX domain
domainId INTEGER, -- User-controlled ID that can be used to group events.
uint64Value INTEGER, -- One of possible payload value union members.
int64Value INTEGER, -- One of possible payload value union members.
doubleValue REAL, -- One of possible payload value union members.
uint32Value INTEGER, -- One of possible payload value union members.
int32Value INTEGER, -- One of possible payload value union members.
floatValue REAL, -- One of possible payload value union members.
jsonTextId INTEGER, -- One of possible payload value union members.
jsonText TEXT -- One of possible payload value union members.
binaryData BLOB -- Binary payload representation. Format: numPayloads|schemaId1|size1|offset1|...|schemaIdN|sizeN|offsetN|data1|...|dataN
);
CREATE TABLE OPENGL_API (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_TYPE(id)
globalTid INTEGER, -- Serialized GlobalId.
endGlobalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- First function name
endNameId INTEGER, -- REFERENCES StringIds(id) -- Last function name
returnValue INTEGER NOT NULL, -- Return value of the function call.
frameId INTEGER, -- Index of the graphics frame starting from 1.
contextId INTEGER, -- Context ID.
gpu INTEGER, -- GPU index.
display INTEGER -- Display ID.
);
CREATE TABLE OPENGL_WORKLOAD (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_TYPE(id)
globalTid INTEGER, -- Serialized GlobalId.
endGlobalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- First function name
endNameId INTEGER, -- REFERENCES StringIds(id) -- Last function name
returnValue INTEGER NOT NULL, -- Return value of the function call.
frameId INTEGER, -- Index of the graphics frame starting from 1.
contextId INTEGER, -- Context ID.
gpu INTEGER, -- GPU index.
display INTEGER -- Display ID.
);
CREATE TABLE KHR_DEBUG_EVENTS (
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_TYPE(id)
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER, -- Event end timestamp (ns).
textId INTEGER, -- REFERENCES StringIds(id) -- Debug marker/group text
globalTid INTEGER, -- Serialized GlobalId.
source INTEGER, -- REFERENCES ENUM_OPENGL_DEBUG_SOURCE(id)
khrdType INTEGER, -- REFERENCES ENUM_OPENGL_DEBUG_TYPE(id)
id INTEGER, -- KHR event ID.
severity INTEGER, -- REFERENCES ENUM_OPENGL_DEBUG_SEVERITY(id)
correlationId INTEGER, -- ID used to correlate KHR CPU trace to GPU trace.
context INTEGER -- Context ID.
);
CREATE TABLE OSRT_API (
-- OS runtime libraries traced to gather information about low-level userspace APIs.
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
returnValue INTEGER NOT NULL, -- Return value of the function call.
nestingLevel INTEGER, -- Zero-base index of the nesting level.
callchainId INTEGER NOT NULL -- REFERENCES OSRT_CALLCHAINS(id)
);
CREATE TABLE OSRT_CALLCHAINS (
-- Callchains attached to OSRT events, depending on selected profiling settings.
id INTEGER NOT NULL, -- Part of PRIMARY KEY (id, stackDepth).
symbol INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
module INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Module name
kernelMode INTEGER, -- True if kernel mode.
thumbCode INTEGER, -- True if thumb code.
unresolved INTEGER, -- True if the symbol was not resolved.
specialEntry INTEGER, -- True if artifical entry added during processing callchain.
originalIP INTEGER, -- Instruction pointer value.
unwindMethod INTEGER, -- REFERENCES ENUM_STACK_UNWIND_METHOD(id)
stackDepth INTEGER NOT NULL, -- Zero-base index of the given function in call stack.
PRIMARY KEY (id, stackDepth)
);
CREATE TABLE PROFILER_OVERHEAD (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
returnValue INTEGER NOT NULL -- Return value of the function call.
);
CREATE TABLE SCHED_EVENTS (
-- Thread scheduling events.
start INTEGER NOT NULL, -- Event start timestamp (ns).
cpu INTEGER NOT NULL, -- ID of CPU this thread was scheduled in or out.
isSchedIn INTEGER NOT NULL, -- 0 if thread was scheduled out, non-zero otherwise.
globalTid INTEGER -- Serialized GlobalId.
threadState INTEGER, -- REFERENCES ENUM_SAMPLING_THREAD_STATE(id)
threadBlock INTEGER, -- REFERENCES ENUM_SCHEDULING_THREAD_BLOCK(id)
);
CREATE TABLE COMPOSITE_EVENTS (
-- Thread sampling events.
id INTEGER NOT NULL PRIMARY KEY, -- ID of the composite event.
start INTEGER NOT NULL, -- Event start timestamp (ns).
cpu INTEGER, -- ID of CPU this thread was running on.
threadState INTEGER, -- REFERENCES ENUM_SAMPLING_THREAD_STATE(id)
globalTid INTEGER, -- Serialized GlobalId.
cpuCycles INTEGER NOT NULL -- Value of Performance Monitoring Unit (PMU) counter.
);
CREATE TABLE SAMPLING_CALLCHAINS (
-- Callchain entries obtained from composite events, used to construct function table views.
id INTEGER NOT NULL, -- REFERENCES COMPOSITE_EVENTS(id)
symbol INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
module INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Module name
kernelMode INTEGER, -- True if kernel mode.
thumbCode INTEGER, -- True if thumb code.
unresolved INTEGER, -- True if the symbol was not resolved.
specialEntry INTEGER, -- True if artifical entry added during processing callchain.
originalIP INTEGER, -- Instruction pointer value.
unwindMethod INTEGER, -- REFERENCES ENUM_STACK_UNWIND_METHOD(id)
stackDepth INTEGER NOT NULL, -- Zero-base index of the given function in call stack.
PRIMARY KEY (id, stackDepth)
);
CREATE TABLE PERF_EVENT_SOC_OR_CPU_RAW_EVENT (
-- SoC and CPU raw event values from Sampled Performance Counters.
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
vmId INTEGER, -- VM ID.
componentId INTEGER, -- REFERENCES TARGET_INFO_COMPONENT(componentId)
eventId INTEGER, -- REFERENCES TARGET_INFO_PERF_METRIC(id)
count INTEGER -- Counter data value
);
CREATE TABLE PERF_EVENT_SOC_OR_CPU_METRIC_EVENT (
-- SoC and CPU metric values from Sampled Performance Counters.
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
vmId INTEGER, -- VM ID.
componentId INTEGER, -- REFERENCES TARGET_INFO_COMPONENT(componentId)
metricId INTEGER, -- REFERENCES TARGET_INFO_PERF_METRIC(id)
value REAL -- Metric data value
);
CREATE TABLE SLI_QUERIES (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
gpu INTEGER NOT NULL, -- GPU index.
frameId INTEGER NOT NULL, -- Index of the graphics frame starting from 1.
occQueryIssued INTEGER NOT NULL, -- Occlusion query issued.
occQueryAsked INTEGER NOT NULL, -- Occlusion query asked.
eventQueryIssued INTEGER NOT NULL, -- Event query issued.
eventQueryAsked INTEGER NOT NULL, -- Event query asked.
numberOfTransferEvents INTEGER NOT NULL, -- Number of transfer events.
amountOfTransferredData INTEGER NOT NULL -- Cumulative size of resource data that was transferred.
);
CREATE TABLE SLI_P2P (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
gpu INTEGER NOT NULL, -- GPU index.
frameId INTEGER NOT NULL, -- Index of the graphics frame starting from 1.
transferSkipped INTEGER NOT NULL, -- Number of transfers that were skipped.
srcGpu INTEGER NOT NULL, -- Source GPU ID.
dstGpu INTEGER NOT NULL, -- Destination GPU ID.
numSubResources INTEGER NOT NULL, -- Number of sub-resources to transfer.
resourceSize INTEGER NOT NULL, -- Size of resource.
subResourceIdx INTEGER NOT NULL, -- Sub-resource index.
smplWidth INTEGER, -- Sub-resource surface width in samples.
smplHeight INTEGER, -- Sub-resource surface height in samples.
smplDepth INTEGER, -- Sub-resource surface depth in samples.
bytesPerElement INTEGER, -- Number of bytes per element.
dxgiFormat INTEGER, -- REFERENCES ENUM_DXGI_FORMAT(id)
logSurfaceNames TEXT, -- Surface name.
transferInfo INTEGER, -- REFERENCES ENUM_SLI_TRANSER(id)
isEarlyPushManagedByNvApi INTEGER, -- True if early push managed by NVAPI. False otherwise.
useAsyncP2pForResolve INTEGER, -- True if async Peer-to-Peer used for resolve. False otherwise.
transferFuncName TEXT, -- "A - BE" for asynchronous transfer, "S - BE" for synchronous transfer.
regimeName TEXT, -- Name of the regime scope that includes the resource.
debugName TEXT, -- Debug name assigned to the resource by the application code.
bindType TEXT -- Bind type.
);
CREATE TABLE SLI_STATS (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
gpu INTEGER NOT NULL, -- GPU index.
countComplexFrames INTEGER NOT NULL, -- Complex frames count.
countStats INTEGER NOT NULL, -- Number of frame statistics collected for the inactive-time histogram.
totalInactiveTime INTEGER NOT NULL, -- Total inactive time (µs).
minPbSize INTEGER NOT NULL, -- Min push buffer size.
maxPbSize INTEGER NOT NULL, -- Max push buffer size.
totalPbSize INTEGER NOT NULL -- Total push buffer size.
);
CREATE TABLE DX12_API (
id INTEGER NOT NULL PRIMARY KEY,
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
shortContextId INTEGER, -- Short form of the COM interface object address.
frameId INTEGER, -- Index of the graphics frame starting from 1.
color INTEGER, -- Encoded ARGB color value.
textId INTEGER, -- REFERENCES StringIds(id) -- PIX marker text
commandListType INTEGER, -- REFERENCES ENUM_D3D12_CMD_LIST_TYPE(id)
objectNameId INTEGER, -- REFERENCES StringIds(id) -- D3D12 object name
longContextId INTEGER -- Long form of the COM interface object address.
);
CREATE TABLE DX12_WORKLOAD (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
shortContextId INTEGER, -- Short form of the COM interface object address.
frameId INTEGER, -- Index of the graphics frame starting from 1.
gpu INTEGER, -- GPU index.
color INTEGER, -- Encoded ARGB color value.
textId INTEGER, -- REFERENCES StringIds(id) -- PIX marker text
commandListType INTEGER, -- REFERENCES ENUM_D3D12_CMD_LIST_TYPE(id)
objectNameId INTEGER, -- REFERENCES StringIds(id) -- D3D12 object name
longContextId INTEGER -- Long form of the COM interface object address.
);
CREATE TABLE DX12_MEMORY_OPERATION (
gpu INTEGER, -- GPU index.
rangeStart INTEGER, -- Offset denoting the beginning of a memory range (B).
rangeEnd INTEGER, -- Offset denoting the end of a memory range (B).
subresourceId INTEGER, -- Subresource index.
heapType INTEGER, -- REFERENCES ENUM_D3D12_HEAP_TYPE(id)
heapFlags INTEGER, -- REFERENCES ENUM_D3D12_HEAP_FLAGS(id)
cpuPageProperty INTEGER, -- REFERENCES ENUM_D3D12_PAGE_PROPERTY(id)
nvApiFlags INTEGER, -- NV specific flags. See docs for specifics.
traceEventId INTEGER NOT NULL -- REFERENCES DX12_API(id)
);
CREATE TABLE VULKAN_API (
id INTEGER NOT NULL PRIMARY KEY,
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
contextId INTEGER -- Short form of the interface object address.
);
CREATE TABLE VULKAN_WORKLOAD (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
gpu INTEGER, -- GPU index.
contextId INTEGER, -- Short form of the interface object address.
color INTEGER, -- Encoded ARGB color value.
textId INTEGER -- REFERENCES StringIds(id) -- Vulkan CPU debug marker string
);
CREATE TABLE VULKAN_DEBUG_API (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
contextId INTEGER, -- Short form of the interface object address.
color INTEGER, -- Encoded ARGB color value.
textId INTEGER -- REFERENCES StringIds(id) -- Vulkan CPU debug marker string
);
CREATE TABLE VULKAN_PIPELINE_CREATION_EVENTS (
id INTEGER NOT NULL PRIMARY KEY, -- ID of the pipeline creation event.
duration INTEGER, -- Event duration (ns).
flags INTEGER, -- REFERENCES ENUM_VULKAN_PIPELINE_CREATION_FLAGS(id)
traceEventId INTEGER NOT NULL -- REFERENCES VULKAN_API(id) -- ID of the attached vulkan API.
);
CREATE TABLE VULKAN_PIPELINE_STAGE_EVENTS (
id INTEGER NOT NULL PRIMARY KEY, -- ID of the pipeline stage event.
duration INTEGER, -- Event duration (ns).
flags INTEGER, -- REFERENCES ENUM_VULKAN_PIPELINE_CREATION_FLAGS(id)
creationEventId INTEGER NOT NULL -- REFERENCES VULKAN_PIPELINE_CREATION_EVENTS(id) -- ID of the attached pipeline creation event.
);
CREATE TABLE GPU_CONTEXT_SWITCH_EVENTS (
tag INTEGER NOT NULL, -- REFERENCES ENUM_GPU_CTX_SWITCH(id)
vmId INTEGER NOT NULL, -- VM ID.
seqNo INTEGER NOT NULL, -- Sequential event number.
contextId INTEGER NOT NULL, -- Context ID.
timestamp INTEGER NOT NULL, -- Event start timestamp (ns).
globalPid INTEGER, -- Serialized GlobalId.
gpuId INTEGER -- GPU index.
);
CREATE TABLE OPENMP_EVENT_KIND_THREAD (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
threadId INTEGER, -- Internal thread sequence starting from 1.
threadType INTEGER -- REFERENCES ENUM_OPENMP_THREAD(id)
);
CREATE TABLE OPENMP_EVENT_KIND_PARALLEL (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- Internal parallel region sequence starting from 1.
parentTaskId INTEGER -- ID for task that creates this parallel region.
);
CREATE TABLE OPENMP_EVENT_KIND_SYNC_REGION_WAIT (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER, -- ID of the task that this event belongs to.
kind INTEGER -- REFERENCES ENUM_OPENMP_SYNC_REGION(id)
);
CREATE TABLE OPENMP_EVENT_KIND_SYNC_REGION (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER, -- ID of the task that this event belongs to.
kind INTEGER -- REFERENCES ENUM_OPENMP_SYNC_REGION(id)
);
CREATE TABLE OPENMP_EVENT_KIND_TASK (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER, -- ID of the task that this event belongs to.
kind INTEGER -- REFERENCES ENUM_OPENMP_TASK_FLAG(id)
);
CREATE TABLE OPENMP_EVENT_KIND_MASTER (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER -- ID of the task that this event belongs to.
);
CREATE TABLE OPENMP_EVENT_KIND_REDUCTION (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER -- ID of the task that this event belongs to.
);
CREATE TABLE OPENMP_EVENT_KIND_TASK_CREATE (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parentTaskId INTEGER, -- ID of the parent task that is creating a new task.
newTaskId INTEGER -- ID of the new task that is being created.
);
CREATE TABLE OPENMP_EVENT_KIND_TASK_SCHEDULE (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
priorTaskId INTEGER, -- ID of the task that is being switched out.
priorTaskStatus INTEGER, -- REFERENCES ENUM_OPENMP_TASK_STATUS(id)
nextTaskId INTEGER -- ID of the task that is being switched in.
);
CREATE TABLE OPENMP_EVENT_KIND_CANCEL (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
taskId INTEGER -- ID of the task that is being cancelled.
);
CREATE TABLE OPENMP_EVENT_KIND_MUTEX_WAIT (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_MUTEX(id)
waitId INTEGER, -- ID indicating the object being waited.
taskId INTEGER -- ID of the task that this event belongs to.
);
CREATE TABLE OPENMP_EVENT_KIND_CRITICAL_SECTION (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_MUTEX(id)
waitId INTEGER -- ID indicating the object being held.
);
CREATE TABLE OPENMP_EVENT_KIND_MUTEX_RELEASED (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_MUTEX(id)
waitId INTEGER, -- ID indicating the object being released.
taskId INTEGER -- ID of the task that this event belongs to.
);
CREATE TABLE OPENMP_EVENT_KIND_LOCK_INIT (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_MUTEX(id)
waitId INTEGER -- ID indicating object being created/destroyed.
);
CREATE TABLE OPENMP_EVENT_KIND_LOCK_DESTROY (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_MUTEX(id)
waitId INTEGER -- ID indicating object being created/destroyed.
);
CREATE TABLE OPENMP_EVENT_KIND_WORKSHARE (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_WORK(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER, -- ID of the task that this event belongs to.
count INTEGER -- Measure of the quantity of work involved in the region.
);
CREATE TABLE OPENMP_EVENT_KIND_DISPATCH (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
kind INTEGER, -- REFERENCES ENUM_OPENMP_DISPATCH(id)
parallelId INTEGER, -- ID of the parallel region that this event belongs to.
taskId INTEGER -- ID of the task that this event belongs to.
);
CREATE TABLE OPENMP_EVENT_KIND_FLUSH (
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
eventClass INTEGER NOT NULL, -- REFERENCES ENUM_NSYS_EVENT_CLASS(id)
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- Currently unused.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Event name
eventKind INTEGER, -- REFERENCES ENUM_OPENMP_EVENT_KIND(id)
threadId INTEGER -- ID of the thread that this event belongs to.
);
CREATE TABLE D3D11_PIX_DEBUG_API (
-- D3D11 debug marker events.
start INTEGER NOT NULL, -- Event start timestamp (ns).
end INTEGER NOT NULL, -- Event end timestamp (ns).
globalTid INTEGER, -- Serialized GlobalId.
correlationId INTEGER, -- First ID matching an API call to GPU workloads.
endCorrelationId INTEGER, -- Last ID matching an API call to GPU workloads.
nameId INTEGER NOT NULL, -- REFERENCES StringIds(id) -- Function name
shortContextId INTEGER, -- Short form of the COM interface object address.
frameId INTEGER,