Usage : nvlink [options] Options ======= --Ofast-compile <0|min|mid|max> (-Ofc) Specify fast-compile level for device code, which controls the tradeoff between compilation speed and runtime performance by disabling certain optimizations at varying levels. Levels: 'max': Focus only on the fastest compilation speed, disabling many optimizations. 'mid': Balance compile time and runtime, disabling expensive optimizations. 'min': More minimal impact on both compile time and runtime, minimizing some expensive optimizations. '0': Disables fast-compile. The option is disabled by default. --Xnvvm ,... (-Xnvvm) Nvvm options (only used with LTO). --Xptxas ,... (-Xptxas) Ptxas options (only used with LTO). --arch (-arch) Specify the 'sm_' name of the target architecture. Allowed values for this option: 'compute_100','compute_100a','compute_100f', 'compute_103','compute_103a','compute_103f','compute_107','compute_107a', 'compute_107f','compute_110','compute_110a','compute_110f','compute_120', 'compute_120a','compute_120f','compute_121','compute_121a','compute_121f', 'compute_75','compute_80','compute_86','compute_87','compute_88','compute_89', 'compute_90','compute_90a','lto_100','lto_100a','lto_100f','lto_103','lto_103a', 'lto_103f','lto_107','lto_107a','lto_107f','lto_110','lto_110a','lto_110f', 'lto_120','lto_120a','lto_120f','lto_121','lto_121a','lto_121f','lto_75', 'lto_80','lto_86','lto_87','lto_88','lto_89','lto_90','lto_90a','sm_100', 'sm_100a','sm_100f','sm_103','sm_103a','sm_103f','sm_107','sm_107a','sm_107f', 'sm_110','sm_110a','sm_110f','sm_120','sm_120a','sm_120f','sm_121','sm_121a', 'sm_121f','sm_75','sm_80','sm_86','sm_87','sm_88','sm_89','sm_90','sm_90a'. --cpu-arch (-cpu-arch) Specify the name of the cpu target architecture. Allowed values for this option: 'AARCH64','ARMv7','PPC64LE','RISCV64','X86', 'X86_64','unknown'. Default value: 'unknown'. --debug (-g) Specify that this was a debug compile. --device-stack-protector (-device-stack-protector) Enable stack protectors. --disable-infos (-disable-infos) Disable all info messages. --disable-warnings (-w) Inhibit all warning messages. --dlto (-dlto) Do link-time optimization, alias for -lto. --dump-callgraph (-dump-callgraph) Dump callgraph information. --dump-callgraph-no-demangle (-dump-callgraph-no-demangle) Dump callgraph information without demangling. --emit-ptx (-emit-ptx) Emit ptx file if LTO is used. --extra-warnings (-extrawarn) Emit extra warnings about possible problems. --gen-host-linker-script (-ghls) Specify the type of host linker script to be generated. Allowed values for this option: 'lcs-abs','lcs-aug'. --help (-h) Print this help information on this tool. --ignore-host-info (-ignore-host-info) Ignore information about host references, so don't remove device code that could potentially be referenced by host. --keep-system-libraries (-keep-system-libraries) Don't optimize away system library (e.g. cudadevrt) code. --kernels-used ,... (-kernels-used) Specify kernels that are used. Can be part of a kernel name so any kernels with that string in name are matched. If this option is used, then any other kernels are considered dead-code and removed. --library ,... (-l) Specify libraries to be used in the linking stage. The libraries are searched for on the library search paths that have been specified using option '-L'. --library-path ,... (-L) Specify library search paths. --link-as-tools-patch (-astoolspatch) Link code compiled with astoolspatch option. --link-time-opt (-lto) Do link-time optimization. --machine (-m) Specify 64 bit architecture. Allowed values for this option: 64. Default value: 64. --maxrregcount (-maxrregcount) Max regcount (only used with LTO). --nvvmpath (-nvvmpath) Path to libnvvm library. --options-file ,... (-optf) Include command line options from specified file. --output-file (-o) Specify name and location of the output file. --preserve-relocs (-preserve-relocs) Preserve resolved relocations in linked executable. --register-link-binaries --register-link-binaries Specify name and location of output file that will contain names of cudaRegister routines for each linked file. --relocatable-link (-r) Do relocatable / incremental link. --report-arch (-report-arch) Report SM target arch in error messages. --split-compile (-split-compile) Max threads nvvm can use for split compilation (only used with LTO). Default value: 1. --split-compile-extended (-split-compile-extended) Max threads linker can use for extended split compilation (only used with LTO). Default value: 1. --suppress-arch-warning (-suppress-arch-warning) Suppress the warning that otherwise is printed when object does not contain code for target arch. --suppress-debug-info (-suppress-debug-info) Do not preserve debug symbols in output. Note: This option is ignored if used without --debug option. --suppress-stack-size-warning (-suppress-stack-size-warning) Suppress the warning that otherwise is printed when stack size cannot be determined. --time (-time) Generate a comma separated value table with the time taken by this program and sub-component invocations, and append it at the end of the file given as the option argument. If the file is empty, the column headings are generated in the first row of the table. If the file name is '-', the timing data is generated in stdout. --uidx-file (-uidx) Path to uidx file. --use-host-info (-use-host-info) Use information about host references to remove unused device code. --variables-used ,... (-variables-used) Specify variables that are used. Can be part of a variable name so any variable with that string in name are matched. If this option is used, then any other variables are considered dead-code and potentially removed unless have other accesses from device code. --verbose (-v) Enable verbose mode which prints code generation statistics. --verbose2 (-v2) Enable extended verbose mode. Extends basic verbose mode with progress indicators, names of processed modules, etc. --version (-V) Print version information on this tool. --warning-as-error (-Werror) Make all warnings into errors.