fn build_kernel()

in crates/ratchet-core/src/ops/matmul/gemm.rs [226:258]

• 33 lines of code
• 1 McCabe index (conditional complexity)

    fn build_kernel(
        &self,
        inplace: bool,
        dst: &Tensor,
        workgroup_size: &WorkgroupSize,
    ) -> Result<KernelSource, OperationError> {
        let kernel_element = self.spec.select_kernel_element();
        match (self.lhs.dt(), kernel_element) {
            (DType::F32, KernelElement::Scalar) => {
                self.render::<Scalar<f32>>(inplace, dst, workgroup_size)
            }
            (DType::F32, KernelElement::Vec2) => {
                self.render::<Vec2<f32>>(inplace, dst, workgroup_size)
            }
            (DType::F32, KernelElement::Vec4) => {
                self.render::<Vec4<f32>>(inplace, dst, workgroup_size)
            }
            (DType::F16, KernelElement::Scalar) => {
                self.render::<Scalar<f16>>(inplace, dst, workgroup_size)
            }
            (DType::F16, KernelElement::Vec2) => {
                self.render::<Vec2<f16>>(inplace, dst, workgroup_size)
            }
            (DType::F16, KernelElement::Vec4) => {
                self.render::<Vec4<f16>>(inplace, dst, workgroup_size)
            }
            (DType::Q8_0F(_), _) => self.render::<Scalar<f32>>(inplace, dst, workgroup_size),
            (DType::Q8_0H(_), _) => self.render::<Scalar<f16>>(inplace, dst, workgroup_size),
            (DType::Q4_KF(_), _) => self.render::<Scalar<f32>>(inplace, dst, workgroup_size),
            (DType::Q4_KH(_), _) => self.render::<Scalar<f16>>(inplace, dst, workgroup_size),
            _ => Err(InvariantError::UnsupportedDType(self.lhs.dt()).into()),
        }
    }