--profile Options - --profile Options - 2026.1 English - UG1702

--profile.aie:<arg>

Enables profiling of AI Engine streams in adaptive data flow (ADF) applications, where <arg> is:

<ADF_graph_argument|pin name|all>

• <ADF_graph_argument>: Specifies an argument name from the ADF graph application.
• <pin_name>: Indicates a port on an AI Engine kernel.
• <all>: Indicates monitoring all stream connections in the ADF application.

v++ --link --profile.aie:DataIn1

--profile.aie_trace_offload:<arg>

Enables profiling of AI Engine streams in adaptive data flow (ADF) applications, where <arg> is HSDP, or high-speed debug port.

v++ --link --profile.aie_trace_offload:HSDP

See the Event Trace Offload using High Speed Debug Port in AI Engine Tools and Flows User Guide (UG1076) for details on the use of HSDP.

--profile.data:<arg>

Enables monitoring of data ports you add to the design via monitor IP. Specify this option during linking.

Where <arg> is:

[<kernel_name>|all]:[<cu_name>|all]:[<interface_name>|all](:[counters|all])

• [<kernel_name>|all] defines a specific kernel to apply the command to. You can also specify the keyword all to apply the monitoring to all existing kernels, compute units, and interfaces with a single option.

• When you specify <kernel_name>, you can target a specific compute unit with <cu_name> or apply the command to all compute units.

• [<interface_name>|all] defines the specific interface on the kernel or CU to monitor for data activity, or monitor all interfaces.
• [<counters|all] is an optional argument, as it defaults to all when not specified. It allows restricting the information gathering to only counters for larger designs, while all will include the collection of actual trace information.

For example, to assign the data profile to all CUs and interfaces of kernel k1 use the following command:

v++ --link --profile.data k1:all:all

--profile.exec:<arg>

This option records the execution times of the kernel and provides minimum port data collection during the system run. This option needs to be specified during linking.

The syntax for exec profiling is:

[<kernel_name>|all]:[<cu_name>|all](:[counters|all])

For example, to profile to execution of cu2 for kernel k1 use the following command:

v++ --link --profile.exec:k1:cu2

--profile.stall:<arg>

Adds stall monitoring logic to the device binary (.xclbin). Requires the addition of stall ports on the kernel interface. You must specify the stall option during both compilation and linking.

The syntax for stall profiling is:

[<kernel_name>|all]:[<cu_name>|all](:[counters|all])

For example, to monitor stalls of cu2 for kernel k1 use the following command:

v++ --compile -k k1 --profile.stall ...
v++ --link --profile.stall:k1:cu2 ...

--profile.trace_memory:<arg>

When building the hardware target (-t=hw), use this option to specify the type and amount of memory to use for capturing trace data. You can specify the argument as follows:

<FIFO>:<size>|<MEMORY>[<n>][:<SLR>]

This argument specifies memory type to use for capturing trace data. Use the --profile.trace_memory command to define the type or memory to use, with the trace_buffer_size switch in the xrt.ini file to define the amount of memory to use as described in xrt.ini File. The default memory type used is the first memory defined in the platform, and the default buffer size is 1 MB.

When you enable device_trace without trace_memory, the tool captures profile data in default platform memory and allocates a 1 MB trace buffer.

FIFO:<size>

Specified in KB. The maximum is 128K, although 64K is the maximum recommended.

Memory

Specifies the type and number of memory resource on the platform. You can identify memory resources for the target platform with the platforminfo command. Supported memory types include HBM, DDR, PLRAM, HP, ACP, MIG, and MC_NOC. For example, DDR[1].

[:<SLR>]

Optional. Indicates that CUs you assign to the <SLR> need to use the DDR or HBM resources in the <MEMORY> field. You can only use this syntax with DDR or HBM memory banks.

You can specify the --profile.trace_memory command with the memory size and unit such as FIFO:8k, or specify the memory bank such as DDR[0] or HBM[3]. In this case, Vitis captures the profile data for all CUs in the specified memory.

You can also specify the memory to use to capture profile data, and the SLR assignment for that memory. In this case, the SLR assignment indicates that any CUs assigned to the specified SLR need to have profile data captured in the specified memory. See the following example.

[profile]
trace_memory=DDR[1]:SLR0
trace_memory=DDR[2]:SLR1

The example captures profile data for CUs on SLR0 in DDR bank 1 and for CUs on SLR1 in DDR bank 2. Use the --connectivity.slr command to assign compute units to SLRs, as described in Assigning Compute Units to SLRs on Alveo Accelerator Cards in the Data Center Acceleration using Vitis (UG1700).