ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-19 12:30:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

Compile for gfx908 and gfx90a (#130)

Browse Source 
* adding compilation for multiple targets

* fix build

* clean

* update Jekinsfile

* update readme

* update Jenkins

* use ck::half_t instead of ushort for bf16

* rename enum classes

* clean

* rename

* clean

[ROCm/composable_kernel commit: cd167e492a]
This commit is contained in:
Chao Liu
2022-03-31 12:33:34 -05:00
committed by GitHub
parent 59506defde
commit 3f732cceab
227 changed files with 1398 additions and 2944 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/include/conv_test_util.hpp
View File

@@ -31,7 +31,7 @@ using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
using OutElementOp = ck::tensor_operation::element_wise::PassThrough;
static constexpr auto ConvFwdDefault =
ck::tensor_operation::device::ConvolutionForwardSpecialization_t::Default;
ck::tensor_operation::device::ConvolutionForwardSpecialization::Default;
template <ck::index_t SpatialDims, typename InDataType, typename WeiDataType, typename OutDataType>
using DeviceConvNDFwdInstance = ck::tensor_operation::device::

2
test/magic_number_division/magic_number_division.cpp
View File

@@ -5,7 +5,7 @@
#include <stdlib.h>
#include <half.hpp>
#include "config.hpp"
#include "print.hpp"
#include "magic_division.hpp"
#include "device.hpp"
#include "host_tensor.hpp"
#include "host_tensor_generator.hpp"

10
test/reduce/reduce_no_index.cpp
View File

@@ -51,11 +51,11 @@ struct type_mapping<ck::half_t>
constexpr int Rank = 4;
constexpr ReduceTensorOp_t ReduceOpId = ReduceTensorOp_t::AVG;
constexpr NanPropagation_t NanOpt = NanPropagation_t::PROPAGATE_NAN;
constexpr bool PropagateNan = false;
constexpr ReduceTensorIndices_t IndicesOpt = ReduceTensorIndices_t::NO_INDICES;
constexpr bool NeedIndices = false;
constexpr ReduceTensorOp ReduceOpId = ReduceTensorOp::AVG;
constexpr NanPropagation NanOpt = NanPropagation::PROPAGATE_NAN;
constexpr bool PropagateNan = false;
constexpr ReduceTensorIndices IndicesOpt = ReduceTensorIndices::NO_INDICES;
constexpr bool NeedIndices = false;
template <typename InDataType,
typename AccDataType,

10
test/reduce/reduce_with_index.cpp
View File

@@ -51,11 +51,11 @@ struct type_mapping<ck::half_t>
constexpr int Rank = 4;
constexpr ReduceTensorOp_t ReduceOpId = ReduceTensorOp_t::AMAX;
constexpr NanPropagation_t NanOpt = NanPropagation_t::PROPAGATE_NAN;
constexpr bool PropagateNan = false;
constexpr ReduceTensorIndices_t IndicesOpt = ReduceTensorIndices_t::FLATTENED_INDICES;
constexpr bool NeedIndices = true;
constexpr ReduceTensorOp ReduceOpId = ReduceTensorOp::AMAX;
constexpr NanPropagation NanOpt = NanPropagation::PROPAGATE_NAN;
constexpr bool PropagateNan = false;
constexpr ReduceTensorIndices IndicesOpt = ReduceTensorIndices::FLATTENED_INDICES;
constexpr bool NeedIndices = true;
template <typename InDataType,
typename AccDataType,