mirror of
https://github.com/ml-explore/mlx.git
synced 2025-07-27 11:31:21 +08:00
cec8661113
* Enable copy to work with int64 strides * Fix uniform buffer indices or copy kernel arguments * Update utils.h * Remove manual unrolling of elem to loc loop * GPU copy updated to handle negative strides * Add slice update primitive
41 lines
899 B
C++
41 lines
899 B
C++
// Copyright © 2023-2024 Apple Inc.
|
|
|
|
#pragma once
|
|
|
|
#include "mlx/backend/common/copy.h"
|
|
#include "mlx/stream.h"
|
|
|
|
namespace mlx::core {
|
|
|
|
// Generic copy inplace
|
|
template <typename stride_t>
|
|
void copy_gpu_inplace(
|
|
const array& in,
|
|
array& out,
|
|
const std::vector<int>& data_shape,
|
|
const std::vector<stride_t>& i_strides,
|
|
const std::vector<stride_t>& o_strides,
|
|
int64_t i_offset,
|
|
int64_t o_offset,
|
|
CopyType ctype,
|
|
const Stream& s);
|
|
|
|
void copy_gpu(const array& src, array& out, CopyType ctype, const Stream& s);
|
|
void copy_gpu(const array& src, array& out, CopyType ctype);
|
|
|
|
void copy_gpu_inplace(
|
|
const array& src,
|
|
array& out,
|
|
CopyType ctype,
|
|
const Stream& s);
|
|
|
|
void copy_gpu_inplace(
|
|
const array& in,
|
|
array& out,
|
|
const std::vector<int64_t>& istride,
|
|
int64_t ioffset,
|
|
CopyType ctype,
|
|
const Stream& s);
|
|
|
|
} // namespace mlx::core
|