Struct GpuInfo 

#[non_exhaustive]
pub struct GpuInfo {
    pub name: String,
    pub vendor: Vendor,
    pub total_bytes: u64,
    pub free_bytes: Option<u64>,
    pub used_bytes: Option<u64>,
    pub arch_target: Option<ArchTarget>,
}

A single detected GPU and its memory.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

name: String

Human-readable name (e.g. "NVIDIA GeForce RTX 4090").

vendor: Vendor

Hardware vendor.

total_bytes: u64

Total memory in bytes. For discrete GPUs this is dedicated VRAM; for integrated/unified GPUs (Intel iGPUs, AMD APUs, Apple Silicon) it is the shared system-memory ceiling available to the GPU, not a dedicated pool.

An AMD APU reports its BIOS VRAM carveout plus the GTT pool the driver allocates from — the latter sized by the kernel’s ttm.pages_limit — since the carveout alone is far below what the part can actually hand out (512 MiB of 14.5 GiB on one such part).

free_bytes: Option<u64>

Free device memory in bytes, when known.

used_bytes: Option<u64>

Used device memory in bytes, when known.

arch_target: Option<ArchTarget>

The architecture a prebuilt artifact must target to run on this GPU: ArchTarget::Gfx on AMD, from KFD sysfs, and ArchTarget::Sm on NVIDIA, from NVML.

None when neither driver reports one — an Apple or Intel GPU, an AMD card on a kernel without KFD, or the nvidia feature disabled. The NVIDIA value also appears on CudaHost, which reports device 0’s alongside the host driver version; this field is per-GPU.

Trait Implementations

impl Clone for GpuInfo

fn clone(&self) -> GpuInfo

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for GpuInfo

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

Formats the value using the given formatter.

impl Display for GpuInfo

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

Formats the value using the given formatter.

impl Eq for GpuInfo

impl PartialEq for GpuInfo

fn eq(&self, other: &GpuInfo) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for GpuInfo