mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-14 02:02:46 +00:00
* updating codegen build for MIOpen access: adding .cmake for codegen component * updating CMake * adding in header guards for some headers due to issues with hiprtc compilation in MIOpen * some more header guards * putting env file in header guard * cleaning up some includes * updated types file for hiprtc purposes * fixed types file: bit-wise/memcpy issue * updating multiple utility files to deal with standard header inclusion for hiprtc * added some more header guards in the utility files, replacing some standard header functionality * added some more header guards * fixing some conflicts in utility files, another round of header guards * fixing errors in data type file * resolved conflict errors in a few utility files * added header guards/replicated functionality in device files * resolved issues with standard headers in device files: device_base and device_grouped_conv_fwd_multiple_abd * resolved issues with standard headers in device files: device_base.hpp, device_grouped_conv_fwd_multiple_abd.hpp, device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp * added header guards for gridwise gemm files: gridwise_gemm_multiple_abd_xdl_cshuffle.hpp and gridwise_gemm_multiple_d_xdl_cshuffle.hpp * fixed issue with numerics header, removed from transform_conv_fwd_to_gemm and added to device_column_to_image_impl, device_grouped_conv_fwd_multiple_abd_xdl_cshuffle, device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3, device_image_to_column_impl * replaced standard header usage and added header guards in block to ctile map and gridwise_gemm_pipeline_selector * resolved errors in device_gemm_xdl_splitk_c_shuffle files in regards to replacement of standard headers in previous commit * added replicated functionality for standard header methods in utility files * replaced standard header functionality in threadwise tensor slice transfer files and added header guards in element_wise_operation.hpp * temp fix for namespace error in MIOpen * remove standard header usage in codegen device op * removed standard header usage in elementwise files, resolved namespace errors * formatting fix * changed codegen argument to ON for testing * temporarily removing codegen compiler flag for testing purposes * added codegen flag again, set default to ON * set codegen flag default back to OFF * replaced enable_if_t standard header usage in data_type.hpp * added some debug prints to pinpoint issues in MIOpen * added print outs to debug in MIOpen * removed debug print outs from device op * resolved stdexcept include error * formatting fix * adding includes to new fp8 file to resolve ck::enable_if_t errors * made changes to amd_wave_read_first_lane * updated functionality in type utility file * fixed end of file issue * resovled errors in type utility file, added functionality to array utility file * fixed standard header usage replication in data_type file, resolves error with failing examples on navi3x * formatting fix * replaced standard header usage in amd_ck_fp8 file * added include to random_gen file * removed and replicated standard header usage from data_type and type_convert files for fp8 changes * replicated standard unsigned integer types in random_gen * resolved comments from review: put calls to reinterpret_cast for size_t in header guards * updated/added copyright headers * removed duplicate header * fixed typo in header guard * updated copyright headers --------- Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
110 lines
3.1 KiB
C++
110 lines
3.1 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2024-2025, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include <rtc/hip.hpp>
|
|
#include <rtc/manage_ptr.hpp>
|
|
#include <stdexcept>
|
|
#include <cassert>
|
|
#include <iostream>
|
|
|
|
namespace rtc {
|
|
|
|
using hip_ptr = RTC_MANAGE_PTR(void, hipFree);
|
|
|
|
std::string hip_error(int error) { return hipGetErrorString(static_cast<hipError_t>(error)); }
|
|
|
|
int get_device_id()
|
|
{
|
|
int device;
|
|
auto status = hipGetDevice(&device);
|
|
if(status != hipSuccess)
|
|
throw std::runtime_error("No device");
|
|
return device;
|
|
}
|
|
|
|
std::string get_device_name()
|
|
{
|
|
hipDeviceProp_t props{};
|
|
auto status = hipGetDeviceProperties(&props, get_device_id());
|
|
if(status != hipSuccess)
|
|
throw std::runtime_error("Failed to get device properties");
|
|
return props.gcnArchName;
|
|
}
|
|
|
|
bool is_device_ptr(const void* ptr)
|
|
{
|
|
hipPointerAttribute_t attr;
|
|
auto status = hipPointerGetAttributes(&attr, ptr);
|
|
if(status != hipSuccess)
|
|
return false;
|
|
return attr.type == hipMemoryTypeDevice;
|
|
}
|
|
|
|
void gpu_sync()
|
|
{
|
|
auto status = hipDeviceSynchronize();
|
|
if(status != hipSuccess)
|
|
throw std::runtime_error("hip device synchronization failed: " + hip_error(status));
|
|
}
|
|
|
|
std::size_t get_available_gpu_memory()
|
|
{
|
|
size_t free;
|
|
size_t total;
|
|
auto status = hipMemGetInfo(&free, &total);
|
|
if(status != hipSuccess)
|
|
{
|
|
std::cerr << "Failed getting available memory: " + hip_error(status) << std::endl;
|
|
return (8ull * 1024ull * 1024ull * 1024ull);
|
|
}
|
|
return free;
|
|
}
|
|
|
|
std::shared_ptr<void> allocate_gpu(std::size_t sz, bool host)
|
|
{
|
|
if(sz > get_available_gpu_memory())
|
|
throw std::runtime_error("Memory not available to allocate buffer: " + std::to_string(sz));
|
|
void* alloc_ptr = nullptr;
|
|
auto status = host ? hipHostMalloc(&alloc_ptr, sz) : hipMalloc(&alloc_ptr, sz);
|
|
if(status != hipSuccess)
|
|
{
|
|
if(host)
|
|
throw std::runtime_error("Gpu allocation failed: " + hip_error(status));
|
|
else
|
|
return allocate_gpu(sz, true);
|
|
}
|
|
assert(alloc_ptr != nullptr);
|
|
std::shared_ptr<void> result = share(hip_ptr{alloc_ptr});
|
|
return result;
|
|
}
|
|
|
|
std::shared_ptr<void> write_to_gpu(const void* x, std::size_t sz, bool host)
|
|
{
|
|
gpu_sync();
|
|
auto result = allocate_gpu(sz, host);
|
|
assert(is_device_ptr(result.get()));
|
|
assert(not is_device_ptr(x));
|
|
auto status = hipMemcpy(result.get(), x, sz, hipMemcpyHostToDevice);
|
|
if(status != hipSuccess)
|
|
throw std::runtime_error("Copy to gpu failed: " + hip_error(status));
|
|
return result;
|
|
}
|
|
|
|
std::shared_ptr<void> read_from_gpu(const void* x, std::size_t sz)
|
|
{
|
|
gpu_sync();
|
|
std::shared_ptr<char> result(new char[sz]);
|
|
assert(not is_device_ptr(result.get()));
|
|
if(not is_device_ptr(x))
|
|
{
|
|
throw std::runtime_error(
|
|
"read_from_gpu() requires Src buffer to be on the GPU, Copy from gpu failed\n");
|
|
}
|
|
auto status = hipMemcpy(result.get(), x, sz, hipMemcpyDeviceToHost);
|
|
if(status != hipSuccess)
|
|
throw std::runtime_error("Copy from gpu failed: " + hip_error(status)); // NOLINT
|
|
return std::static_pointer_cast<void>(result);
|
|
}
|
|
|
|
} // namespace rtc
|