Struct NcclBuildArgs 

pub struct NcclBuildArgs {
    pub tag: Option<String>,
    pub archs: Option<String>,
    pub jobs: usize,
    pub dry_run: bool,
}

Build NCCL from NVIDIA source for a heterogeneous-arch rig.

Fields

tag: Option<String>

NCCL git tag to build (e.g. “v2.27.5-1”). Default: infer from the active libtorch’s bundled NCCL version string (the version we must match for cross-rank handshake).

archs: Option<String>

Override CUDA architectures (semicolon-separated, e.g. “6.1;12.0”). Default: auto-detect from local GPUs.

jobs: usize

Parallel compilation jobs.

dry_run: bool

Show what would happen without building.

Trait Implementations

impl Debug for NcclBuildArgs

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl FdlArgsTrait for NcclBuildArgs

fn try_parse_from(args: &[String]) -> Result<Self, String>

Parse from an explicit argv slice. First element is the program name (ignored), following elements are flags/values/positionals.

fn schema() -> Schema

Return the JSON schema for this CLI shape.

fn render_help() -> String

Render --help to a string.

fn parse() -> Self

Parse argv into Self. Uses std::env::args() by default.

fn render_help_path(argv: &[String]) -> String

Render --help for a specific argv path (program name first, then the tokens typed). The default ignores the path and returns Self::render_help — correct for a single struct, whose help is context-free.