Skip to content

GitLab

Commit b7a6f810 authored by Chao Liu's avatar Chao Liu
Browse files

Merge remote-tracking branch 'origin/develop' into fix_threadwise_copy_error_in_reduction

parents b29dfd70 827301d9
Hide whitespace changes
Inline Side-by-side
Showing with 317 additions and 317 deletions
library/include/ck/library/tensor_operation_instance/gpu/reduce/device_reduce_instance_threadwise_f32_f32_f32.hpp
View file @ b7a6f810
......@@ -11,34 +11,34 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 0, 0, 0, 4, 3); // for ADD
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 0, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 0, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 5, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 5, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 7, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 7, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 1, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 1, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 1, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 1, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 1, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 1, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 5, 0, 0, 4, 3); // for AVG
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 5, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 5, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 7, 0, 0, 4, 3); // for NORM2
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 7, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 7, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 0, 4, 3); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 0, 4, 3); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 0, 4, 3); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 1, 4, 3); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 1, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 2, 0, 1, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 1, 4, 3); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 1, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 3, 0, 1, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 1, 4, 3); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 1, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, float, float, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/include/ck/library/tensor_operation_instance/gpu/reduce/device_reduce_instance_threadwise_f32_f64_f32.hpp
View file @ b7a6f810
......@@ -11,16 +11,16 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 0, 0, 0, 4, 3); // for ADD
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 0, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 0, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 5, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 5, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 7, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 7, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 5, 0, 0, 4, 3); // for AVG
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 5, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 5, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 7, 0, 0, 4, 3); // for NORM2
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 7, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(float, double, float, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/include/ck/library/tensor_operation_instance/gpu/reduce/device_reduce_instance_threadwise_f64_f64_f64.hpp
View file @ b7a6f810
......@@ -11,34 +11,34 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 0, 0, 0, 4, 3); // for ADD
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 0, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 0, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 5, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 5, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 7, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 7, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 0, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 0, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 1, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 1, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 1, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 1, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 1, 4, 0); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 1, 2, 1); //
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 5, 0, 0, 4, 3); // for AVG
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 5, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 5, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 7, 0, 0, 4, 3); // for NORM2
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 7, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 7, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 0, 4, 3); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 0, 4, 3); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 0, 4, 3); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 0, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 0, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 1, 4, 3); // for MIN
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 1, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 2, 0, 1, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 1, 4, 3); // for MAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 1, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 3, 0, 1, 2, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 1, 4, 3); // for AMAX
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 1, 4, 1);
ADD_THREADWISE_INST_REF_BY_ID(double, double, double, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_f16_f16_f16.cpp
View file @ b7a6f810
......@@ -6,25 +6,25 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 2, 1); //
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 3); // for MIN
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 3); // for MAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 3); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 3); // for MIN
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 3); // for MAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 3); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_f16_f32_f16.cpp
View file @ b7a6f810
......@@ -6,16 +6,16 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_f32_f32_f32.cpp
View file @ b7a6f810
......@@ -6,34 +6,34 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 7, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 3); // for MIN
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 3); // for MAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 3); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 3); // for MIN
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 2, 0, 1, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 3); // for MAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 3, 0, 1, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 3); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, float, float, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_f32_f64_f32.cpp
View file @ b7a6f810
......@@ -6,16 +6,16 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(float, double, float, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_f64_f64_f64.cpp
View file @ b7a6f810
......@@ -6,34 +6,34 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 0, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 0, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 1, 4, 0); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 1, 2, 1); //
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 7, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 0, 4, 3); // for MIN
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 0, 4, 3); // for MAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 0, 4, 3); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 0, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 0, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 1, 4, 3); // for MIN
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 2, 0, 1, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 1, 4, 3); // for MAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 3, 0, 1, 2, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 1, 4, 3); // for AMAX
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 1, 4, 1);
ADD_BLOCKWISE_INST_BY_ID(double, double, double, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_second_call_f16_f16_f16.cpp
View file @ b7a6f810
......@@ -6,25 +6,25 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 2, 1); //
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 3); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 3); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 3); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 3); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 3); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 3); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_second_call_f32_f32_f16.cpp
View file @ b7a6f810
......@@ -6,16 +6,16 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, half_t, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_second_call_f32_f32_f32.cpp
View file @ b7a6f810
......@@ -6,34 +6,34 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 7, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 0, 4, 3); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 0, 4, 3); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 0, 4, 3); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 1, 4, 3); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 2, 0, 1, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 1, 4, 3); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 3, 0, 1, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 1, 4, 3); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(float, float, float, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_second_call_f64_f64_f32.cpp
View file @ b7a6f810
......@@ -6,16 +6,16 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, float, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_blockwise_second_call_f64_f64_f64.cpp
View file @ b7a6f810
......@@ -6,34 +6,34 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 0, 0, 0, 4, 3); // for ADD
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 0, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 0, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 5, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 5, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 7, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 7, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 0, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 0, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 1, 4, 0); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 1, 2, 1); //
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 5, 0, 0, 4, 3); // for AVG
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 5, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 5, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 7, 0, 0, 4, 3); // for NORM2
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 7, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 7, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 0, 4, 3); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 0, 4, 3); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 0, 4, 3); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 0, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 0, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 1, 4, 3); // for MIN
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 2, 0, 1, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 1, 4, 3); // for MAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 3, 0, 1, 2, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 1, 4, 3); // for AMAX
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 1, 4, 1);
ADD_BLOCKWISE_SECOND_CALL_INST_BY_ID(double, double, double, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_atomic_add_f16_f32_f32.cpp
View file @ b7a6f810
......@@ -6,13 +6,13 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 0, 0, 0, 4, 3); // for ADD
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 0, 0, 0, 4, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 0, 0, 0, 2, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 5, 0, 0, 4, 0); //
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 5, 0, 0, 2, 1); //
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 5, 0, 0, 4, 3); // for AVG
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 5, 0, 0, 4, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(half_t, float, float, 5, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_atomic_add_f32_f32_f32.cpp
View file @ b7a6f810
......@@ -6,13 +6,13 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 0, 0, 0, 4, 3); // for ADD
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 0, 0, 0, 4, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 0, 0, 0, 2, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 5, 0, 0, 4, 0); //
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 5, 0, 0, 2, 1); //
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 5, 0, 0, 4, 3); // for AVG
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 5, 0, 0, 4, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, float, float, 5, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_atomic_add_f32_f64_f32.cpp
View file @ b7a6f810
......@@ -6,13 +6,13 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 0, 0, 0, 4, 3); // for ADD
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 0, 0, 0, 4, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 0, 0, 0, 2, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 5, 0, 0, 4, 0); //
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 5, 0, 0, 2, 1); //
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 5, 0, 0, 4, 3); // for AVG
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 5, 0, 0, 4, 1);
ADD_MULTIBLOCK_ATOMIC_ADD_INST_BY_ID(float, double, float, 5, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_partial_reduce_f16_f16_f16.cpp
View file @ b7a6f810
......@@ -6,25 +6,25 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 2, 1); //
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 3); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 3); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 3); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 3); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 2, 0, 1, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 3); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 3, 0, 1, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 3); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, half_t, half_t, 4, 0, 1, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_partial_reduce_f16_f32_f16.cpp
View file @ b7a6f810
......@@ -6,16 +6,16 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 0, 1, 2); // for ADD
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 0);
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 3); // for ADD
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 0, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 0, 1, 2); // for AVG
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 3); // for AVG
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 5, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 3); // for NORM2
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(half_t, float, half_t, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_partial_reduce_f32_f32_f32.cpp
View file @ b7a6f810
......@@ -6,29 +6,29 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 0, 1, 2); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 0, 1, 2); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 0, 1, 2); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 0, 1, 2); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 1, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 0, 1, 2); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 1, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 0, 1, 2); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 1, 2, 1); //
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 3); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 3); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 3); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 0, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 3); // for MIN
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 1, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 2, 0, 1, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 3); // for MAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 1, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 3, 0, 1, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 3); // for AMAX
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 1, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 4, 0, 1, 2, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 7, 0, 0, 2, 1); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 3); // for NORM2
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 7, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, float, float, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
library/src/tensor_operation_instance/gpu/reduce/device_reduce_instance_multiblock_partial_reduce_f32_f64_f32.cpp
View file @ b7a6f810
......@@ -6,10 +6,10 @@ namespace device {
namespace device_reduce_instance {
// clang-format off
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | ReduceDims
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 0, 1, 2); // for NORM2
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 0); //
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, double, float, 7, 0, 0, 2, 1); //
// InDataType | AccDataType | OutDataType | ReduceOpId | NanPropaOpt | IndicesOpt | Rank | NumReduceDim
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 3); // for NORM2
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, double, float, 7, 0, 0, 4, 1);
ADD_MULTIBLOCK_PARTIAL_REDUCE_INST_BY_ID(float, double, float, 7, 0, 0, 2, 1);
// clang-format on
} // namespace device_reduce_instance
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment