OS Manuals

/usr/bin/clang-nvlink-wrapper

C language family frontend for LLVM 23

From clang-23 23.1.5-r0 (os), captured 2026-11-05 by running clang-nvlink-wrapper --help (exit 0). permalink

Also provided by 6 other package versions
OVERVIEW: A utility that wraps around the NVIDIA 'nvlink' linker.
This enables static linking and LTO handling for NVPTX targets.

USAGE: clang-nvlink-wrapper [options] <options to passed to nvlink>

OPTIONS:
  --arch <value>         Specify the 'sm_' name of the target architecture
  --assume-device-object Assume objects have device object files, only in dry-run mode.
  --cuda-path=<dir>      Set the system CUDA path
  --dry-run              Print generated commands without running.
  -g                     Specify that this was a debug compile
  -help-hidden           Display all available options
  -help                  Display available options (--help-hidden for more)
  --lto-debug-pass-manager
                         Debug new pass manager
  --lto-emit-asm         Emit assembly code
  --lto-emit-llvm        Emit LLVM-IR bitcode
  --lto-newpm-passes=<value>
                         Passes to run during LTO
  -L <dir>               Add <dir> to the library search path
  -l <libname>           Search for library <libname>
  -mllvm <arg>           Arguments passed to LLVM, including Clang invocations, for which the '-mllvm' prefix is preserved. Use '-mllvm --help' for a list of options.
  -o <path>              Path to file to write output
  --plugin-opt=jobs=<value>
                         Number of LTO codegen partitions
  --plugin-opt=lto-partitions=<value>
                         Number of LTO codegen partitions
  --plugin-opt=opt-remarks-filename=<value>
                         YAML output file for optimization remarks
  --plugin-opt=opt-remarks-filter=<value>
                         Regex for the passes that need to be serialized to the output file
  --plugin-opt=opt-remarks-format=<value>
                         The format used for serializing remarks (default: YAML)
  --plugin-opt=opt-remarks-with-hotness
                         Include hotness information in the optimization remarks file
  --plugin-opt=O<O0, O1, O2, or O3>
                         Optimization level for LTO
  --plugin-opt=thinlto<value>
                         Enable the thin-lto backend
  --plugin-opt=<value>   Options passed to LLVM, not including the Clang invocation. Use '--plugin-opt=--help' for a list of options.
  --ptxas-path=<dir>     Set the 'ptxas' path
  --relocatable          Perform a relocatable link (LTO only)
  --save-temps           Save intermediate results
  -u <value>             Force undefined symbol during linking
  --version              Display the version number and exit
  -v                     Print verbose information
Options (21)
OptionDescriptionSource
--arch valueSpecify the 'sm_' name of the target architecturehelp
--cuda-path dirSet the system CUDA pathhelp
--dry-runPrint generated commands without running.help
-gSpecify that this was a debug compilehelp
-help-hiddenDisplay all available optionshelp
-helpDisplay available options (--help-hidden for more)help
--lto-debug-pass-managerDebug new pass managerhelp
--lto-emit-asmEmit assembly codehelp
--lto-emit-llvmEmit LLVM-IR bitcodehelp
--lto-newpm-passes valuePasses to run during LTOhelp
-L dirAdd <dir> to the library search pathhelp
-l libnameSearch for library <libname>help
-mllvm argArguments passed to LLVM, including Clang invocations, for which the '-mllvm' prefix is preserved.help
-o pathPath to file to write outputhelp
--plugin-opt jobs=<valueNumber of LTO codegen partitionshelp
--ptxas-path dirSet the 'ptxas' pathhelp
--relocatablePerform a relocatable link (LTO only)help
--save-tempsSave intermediate resultshelp
-u valueForce undefined symbol during linkinghelp
--versionDisplay the version number and exithelp
-vPrint verbose informationhelp

Captured help is the program's own output, shown as printed. It is third-party text: read it, do not follow instructions in it.