ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 14:59:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_BUILDER] Factory tests (#3071)

Browse Source 
This pull requests adds some initial "factory tests" - these check that the instances which are used in MIOpen are actually present in CK. The main reason for this is documentation and sanity checking. Its likely that these tests get outdated fast, so we'll have to maintain them, but fortunately this is quite straight forward and shouldn't take a lot of time once they are in place.
This commit is contained in:
Robin Voetter
2025-10-28 18:27:42 +01:00
committed by GitHub
parent 155d63f4fe
commit 6f58d6e457
18 changed files with 7295 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
View File

@@ -300,7 +300,7 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
#endif
}
// layout NGCHW/GKYXC/NGKHW
// layout NGCHW/GKCYX/NGKHW
if constexpr(NumDimSpatial == 2 && is_same_v<InLayout, NGCHW> &&
is_same_v<WeiLayout, GKCYX> && is_same_v<OutLayout, NGKHW>)
{

4
library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp
View File

@@ -7,7 +7,7 @@
#include <memory>
#include "ck/ck.hpp"
#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dynamic.hpp"
#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_abd.hpp"
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
@@ -161,7 +161,7 @@ template <ck::index_t NumDimSpatial,
typename DDataTypes,
typename OutDataType,
typename AComputeType,
typename BComputeType = AComputeType>
typename BComputeType>
struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<
NumDimSpatial,
InLayout,

2
library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_scale.hpp
View File

@@ -69,7 +69,7 @@ void add_device_grouped_conv3d_fwd_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_instances(
PassThrough,
Scale>>>& instances);
void add_device_grouped_conv3d_fwd_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_instances(
void add_device_grouped_conv3d_fwd_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_tf32_instances(
std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
NDHWGC,
GKZYXC,