XeTLA v0.3.6
IntelĀ® Xe Templates for Linear Algebra - API Definition Document
 
Loading...
Searching...
No Matches
gpu::xetla::group::compute_attr_t< dtype_a_, dtype_b_, dtype_acc_ > Struct Template Reference

Compute attribute for gemm. More...

#include <common.hpp>

Public Types

using dtype_a = dtype_a_
 
using dtype_b = dtype_b_
 
using dtype_acc = dtype_acc_
 

Detailed Description

template<typename dtype_a_, typename dtype_b_, typename dtype_acc_>
struct gpu::xetla::group::compute_attr_t< dtype_a_, dtype_b_, dtype_acc_ >

Compute attribute for gemm.

Template Parameters
dtype_a_Is the memory data type of matA.
dtype_b_Is the memory data type of matB.
dtype_acc_Is the compute data type.

Member Typedef Documentation

◆ dtype_a

template<typename dtype_a_ , typename dtype_b_ , typename dtype_acc_ >
using gpu::xetla::group::compute_attr_t< dtype_a_, dtype_b_, dtype_acc_ >::dtype_a = dtype_a_

◆ dtype_acc

template<typename dtype_a_ , typename dtype_b_ , typename dtype_acc_ >
using gpu::xetla::group::compute_attr_t< dtype_a_, dtype_b_, dtype_acc_ >::dtype_acc = dtype_acc_

◆ dtype_b

template<typename dtype_a_ , typename dtype_b_ , typename dtype_acc_ >
using gpu::xetla::group::compute_attr_t< dtype_a_, dtype_b_, dtype_acc_ >::dtype_b = dtype_b_