Module Specific Options - 2025.2 English - UG1702

Options that apply to kernel implementation on AI Engine tiles.

Xchess arg

Passes kernel specific options to the AI Engine compiler, which in turn uses the CHESS compiler to compile code for each AI Engine. Specify the option with the string <kernel>:<optionid>=<value>. The compilation of source files on the kernel mapped AI Engine includes the option string.

Xchess=main:darts.xargs=-nb

Xelfgen arg

Passes additional command-line options to the ELF generation phase of the compiler. Currently, the options run as a make command to build all AI Engine ELF files. Provide the values for this option in quotes, useful if you want to pass special characters. For example, Xelfgen="CLANG_OPTS='-g1 -mllvm --issue-limit=6".

For example, to limit the number of parallel compilations to four, use Xelfgen="-j4".

Note: If errors with bad_alloc appear in the log, or if the Vitis IDE crashes, it can be due to insufficient memory on your workstation. A possible workaround is to limit the parallelism used by the compiler during code generation. Pass this option to the Elf Generator Makefile (-j1 or -j2).

Xelfgen=-j2

Xmapper arg

Passes additional command-line options to the mapper phase of the compiler. Useful for when the design is either failing to converge in the mapping or routing phase, or for better performance via reduction in memory bank conflict.

Xmapper=DisableFloorplanning

Xpreproc arg

Passes general options to the PREPROCESSOR phase for all source code compilations AIE/PS/PL/x86sim of the AI Engine component (e.g., -D...).

Xpreproc=-D<var>=<value>

Xpslinker arg

Pass general option to the PS LINKER phase of the AI Engine component (e.g., -L... -l...).

Xpslinker=-L<libpath> -l<libname>

Xrouter arg

Pass general option to the ROUTER phase.

Xrouter=dmaFIFOsInFreeBankOnly

Xx86sim arg

Pass x86sim-specific option to the compiler.

Xx86sim=clangStaticAnalyzer

fast-floats

Enable fast implementation for linear floating point scalar operations like add, sub, mul, and compare. Accepted values are true and false. The default is false.

fast-floats=true

fast-nonlinearfloats

Enable fast implementation for non-linear floating point scalar operations like sine/cosine, sqrt, and inv. Accepted values are true and false. The default is false.

fast-nonlinearfloats=true

fastmath

Enable fast implementations of float2fix, fplt, and fpge. Accepted values are true and false. The default is false.

fastmath=true

float-accuracy

Option available only for AI Engine-ML. It selects the required floating-point compute accuracy emulated with multiple bfloat16 numerical type operations.

float-accuracy=<arg>

Available argument values are as follows:

• safe: Accuracy is slightly better than single precision floating-point (FP32).
• fast: Improved performance with similar accuracy to FP32 (default).
• low: Best performance with better accuracy than FP16 and bfloat16.