Struct GpuContext 

pub struct GpuContext { /* private fields */ }

GPU compute context for geometric algebra operations.

Manages WebGPU device, queue, and compute pipelines for parallel geometric algebra computation.

Implementations

impl GpuContext

pub async fn new() -> Result<Self, GpuError>

Create a new GPU context.

This initializes WebGPU and creates all compute pipelines.

pub async fn batch_geometric_product( &self, a: &[GA3], b: &[GA3], ) -> Result<Vec<GA3>, GpuError>

Batch geometric product: a[i] * b[i] for all i.

Computes the geometric product of corresponding elements from two input arrays in parallel on the GPU.

pub async fn batch_addition( &self, a: &[GA3], b: &[GA3], ) -> Result<Vec<GA3>, GpuError>

Batch addition: a[i] + b[i] for all i.

pub async fn batch_sandwich( &self, rotors: &[GA3], vectors: &[GA3], ) -> Result<Vec<GA3>, GpuError>

Batch sandwich product: rotor[i] * x[i] * ~rotor[i] for all i.

The sandwich product applies a rotation to each element.

pub async fn batch_exp(&self, a: &[GA3]) -> Result<Vec<GA3>, GpuError>

Batch exponential: exp(a[i]) for all i.

The exponential map converts bivectors to rotors.

pub async fn batch_rotor_slerp( &self, a: &[GA3], b: &[GA3], t: f32, ) -> Result<Vec<GA3>, GpuError>

Batch rotor SLERP: interpolate from a[i] to b[i] by t.

Spherical linear interpolation for smooth rotation blending.

pub fn should_use_gpu(&self, batch_size: usize) -> bool

Check if GPU dispatch is recommended for the given batch size.

pub fn device_info(&self) -> String

Get the device info for debugging.