COMPMID-1698: Implementing CLGEMMLowpMatrixMultiplyReshapedKernel
Change-Id: Ia4db21b394a0b9235393202ce3c00b11cceb94ea
Reviewed-on: https://review.mlplatform.org/568
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
diff --git a/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp b/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
index 4b72878..2a01db7 100644
--- a/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
+++ b/src/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.cpp
@@ -31,43 +31,25 @@
#include "arm_compute/core/Validate.h"
#include "arm_compute/core/utils/misc/ShapeCalculator.h"
#include "arm_compute/runtime/CL/CLScheduler.h"
+#include "arm_compute/runtime/CL/gemm_reshaped/CLGEMMReshapedConfiguration.h"
namespace arm_compute
{
using namespace arm_compute::misc::shape_calculator;
+using namespace arm_compute::cl_gemm;
namespace
{
-inline bool is_interleaved_transposed(int m, int n, int k, bool reshape_b_only_on_first_run, GPUTarget gpu_target)
+inline bool is_gemm_reshaped(unsigned int m, bool reshape_b_only_on_first_run, GPUTarget gpu_target)
{
- bool flag = true;
-
- if(gpu_target_is_in(gpu_target,
- GPUTarget::G71, GPUTarget::G72,
- GPUTarget::G51, GPUTarget::G51BIG, GPUTarget::G51LIT))
- {
- // COMPMID-852
- if(k > 256 && m > 4 && reshape_b_only_on_first_run)
- {
- flag = ((0.72f + n * 0.10766f) < (n * 0.1284f));
- }
- else
- {
- flag = false;
- }
- }
- else
- {
- flag = m > 1;
- }
-
- return flag;
+ return (get_arch_from_target(gpu_target) != GPUTarget::MIDGARD) && (m > 1) && (reshape_b_only_on_first_run);
}
} // namespace
CLGEMMLowpMatrixMultiplyCore::CLGEMMLowpMatrixMultiplyCore(std::shared_ptr<IMemoryManager> memory_manager)
: _memory_group(std::move(memory_manager)),
_mm_kernel(),
+ _mm_reshaped_kernel(),
_mtx_a_reshape_kernel(),
_mtx_b_reshape_kernel(),
_mtx_a_reduction_kernel(),
@@ -82,7 +64,7 @@
_original_b(nullptr),
_a_offset(0),
_b_offset(0),
- _is_interleaved_transposed(true),
+ _is_gemm_reshaped(true),
_reshape_b_only_on_first_run(false),
_is_prepared(false),
_fuse_output_stage(false)
@@ -115,29 +97,17 @@
// Arguments used by GEMMReshapeInfo
// If we pass the matrix A and matrix B reshaped to CLGEMMMatrixMultiplyKernel, we need to pass m, n, k, mult_transpose1xW_width and mult_interleave4x4_height to CLGEMMReshapeInfo
// in order to know how the matrices have been reshaped
- bool reinterpret_input_as_3d = gemm_info.reinterpret_input_as_3d();
- const bool unroll_block = dot8_supported(CLKernelLibrary::get().get_device());
- const int m = reinterpret_input_as_3d ? (a->info()->dimension(1) * a->info()->dimension(2)) : a->info()->dimension(1);
- const int n = b->info()->dimension(0);
- const int k = a->info()->dimension(0);
- const int depth_output_gemm3d = gemm_info.depth_output_gemm3d();
- constexpr int mult_transpose1xW_width = 1;
- constexpr int mult_interleave4x4_height = 1;
- rhs_info.n0 = 16 / b->info()->element_size();
- rhs_info.k0 = 1;
- rhs_info.h0 = mult_transpose1xW_width;
- rhs_info.interleave = false;
- rhs_info.transpose = false;
- lhs_info.m0 = 4;
- lhs_info.k0 = 4;
- lhs_info.v0 = mult_interleave4x4_height;
- lhs_info.interleave = true;
- lhs_info.transpose = !unroll_block;
+ bool reinterpret_input_as_3d = gemm_info.reinterpret_input_as_3d();
+ const unsigned int m = reinterpret_input_as_3d ? (a->info()->dimension(1) * a->info()->dimension(2)) : a->info()->dimension(1);
+ const unsigned int n = b->info()->dimension(0);
+ const unsigned int k = a->info()->dimension(0);
+ const unsigned int batch_size = reinterpret_input_as_3d ? a->info()->dimension(3) : a->info()->dimension(2);
+ const int depth_output_gemm3d = gemm_info.depth_output_gemm3d();
// Check if we need to reshape the matrix A and matrix B
- _is_interleaved_transposed = is_interleaved_transposed(m, n, k, _reshape_b_only_on_first_run, gpu_target);
+ _is_gemm_reshaped = is_gemm_reshaped(m, _reshape_b_only_on_first_run, gpu_target);
- if(_is_interleaved_transposed)
+ if(_is_gemm_reshaped)
{
// if _is_interleaved_transposed is set, force reinterpret_input_as_3d to be false as the output of CLGEMMInterleaveKernel will be 2D
reinterpret_input_as_3d = false;
@@ -151,6 +121,9 @@
_memory_group.manage(&_tmp_b);
}
+ // Pick up the GEMM configuration
+ std::tie(lhs_info, rhs_info) = CLGEMMReshapedConfigurationFactory::create()->configure(m, n, k, batch_size, DataType::QASYMM8);
+
// Configure interleave kernel
_mtx_a_reshape_kernel.configure(a, &_tmp_a, lhs_info, gemm_info.reinterpret_input_as_3d());
@@ -190,10 +163,16 @@
_memory_group.manage(&_mm_result_s32);
- // Configure matrix multiply kernel
- _mm_kernel.configure(matrix_a, matrix_b, &_mm_result_s32, _is_interleaved_transposed, GEMMReshapeInfo(m, n, k,
- mult_transpose1xW_width, mult_interleave4x4_height,
- depth_output_gemm3d, reinterpret_input_as_3d));
+ if(_is_gemm_reshaped)
+ {
+ // Configure and tune matrix multiply kernel
+ _mm_reshaped_kernel.configure(matrix_a, matrix_b, &_mm_result_s32, lhs_info, rhs_info, GEMMReshapeInfo(m, n, k, 1, 1, depth_output_gemm3d, reinterpret_input_as_3d));
+ }
+ else
+ {
+ // Configure matrix multiply kernel
+ _mm_kernel.configure(matrix_a, matrix_b, &_mm_result_s32, false, GEMMReshapeInfo(m, n, k, 1, 1, depth_output_gemm3d, reinterpret_input_as_3d));
+ }
// Configure offset contribution kernel
_offset_contribution_output_stage_kernel.configure(&_mm_result_s32, _a_offset == 0 ? nullptr : &_vector_sum_col, _b_offset == 0 ? nullptr : &_vector_sum_row, c, output, a->info()->dimension(0),
@@ -203,17 +182,23 @@
}
else
{
- // Configure matrix multiply kernel
- _mm_kernel.configure(matrix_a, matrix_b, output, _is_interleaved_transposed, GEMMReshapeInfo(m, n, k,
- mult_transpose1xW_width, mult_interleave4x4_height,
- depth_output_gemm3d, reinterpret_input_as_3d));
+ if(_is_gemm_reshaped)
+ {
+ // Configure and tune matrix multiply kernel
+ _mm_reshaped_kernel.configure(matrix_a, matrix_b, output, lhs_info, rhs_info, GEMMReshapeInfo(m, n, k, 1, 1, depth_output_gemm3d, reinterpret_input_as_3d));
+ }
+ else
+ {
+ // Configure matrix multiply kernel
+ _mm_kernel.configure(matrix_a, matrix_b, output, false, GEMMReshapeInfo(m, n, k, 1, 1, depth_output_gemm3d, reinterpret_input_as_3d));
+ }
// Configure offset contribution kernel
_offset_contribution_kernel.configure(output, _a_offset == 0 ? nullptr : &_vector_sum_col, _b_offset == 0 ? nullptr : &_vector_sum_row, c, a->info()->dimension(0), _a_offset, _b_offset);
}
// Allocate tensors
- if(_is_interleaved_transposed)
+ if(_is_gemm_reshaped)
{
_tmp_a.allocator()->allocate();
if(!_reshape_b_only_on_first_run)
@@ -251,26 +236,14 @@
GEMMRHSMatrixInfo rhs_info;
GEMMLHSMatrixInfo lhs_info;
- bool reinterpret_input_as_3d = gemm_info.reinterpret_input_as_3d();
- const bool unroll_block = dot8_supported(CLKernelLibrary::get().get_device());
- const int m = reinterpret_input_as_3d ? (a->dimension(1) * a->dimension(2)) : a->dimension(1);
- const int n = b->dimension(0);
- const int k = a->dimension(0);
- constexpr int mult_transpose1xW_width = 1;
- constexpr int mult_interleave4x4_height = 1;
- const int depth_output_gemm3d = gemm_info.depth_output_gemm3d();
- rhs_info.n0 = 16 / b->element_size();
- rhs_info.k0 = 1;
- rhs_info.h0 = mult_transpose1xW_width;
- rhs_info.interleave = false;
- rhs_info.transpose = false;
- lhs_info.m0 = 4;
- lhs_info.k0 = 4;
- lhs_info.v0 = mult_interleave4x4_height;
- lhs_info.interleave = true;
- lhs_info.transpose = !unroll_block;
+ bool reinterpret_input_as_3d = gemm_info.reinterpret_input_as_3d();
+ const unsigned int m = reinterpret_input_as_3d ? (a->dimension(1) * a->dimension(2)) : a->dimension(1);
+ const unsigned int n = b->dimension(0);
+ const unsigned int k = a->dimension(0);
+ const unsigned int batch_size = reinterpret_input_as_3d ? a->dimension(3) : a->dimension(2);
+ const int depth_output_gemm3d = gemm_info.depth_output_gemm3d();
- bool reshape_matrices = is_interleaved_transposed(m, n, k, gemm_info.reshape_b_only_on_first_run(), CLScheduler::get().target());
+ bool reshape_matrices = is_gemm_reshaped(m, gemm_info.reshape_b_only_on_first_run(), CLScheduler::get().target());
// if reshape_matrices is set, force reinterpret_input_as_3d to be false as the output of CLGEMMInterleaveKernel will be 2D
if(reshape_matrices)
@@ -278,13 +251,16 @@
reinterpret_input_as_3d = false;
}
- const GEMMReshapeInfo reshape_info = GEMMReshapeInfo(m, n, k, mult_transpose1xW_width, mult_interleave4x4_height, depth_output_gemm3d, reinterpret_input_as_3d);
+ const GEMMReshapeInfo reshape_info = GEMMReshapeInfo(m, n, k, 1, 1, depth_output_gemm3d, reinterpret_input_as_3d);
if(reshape_matrices)
{
matrix_a_info = &tmp_a_info;
matrix_b_info = &tmp_b_info;
+ // Pick up the GEMM configuration
+ std::tie(lhs_info, rhs_info) = CLGEMMReshapedConfigurationFactory::create()->configure(m, n, k, batch_size, DataType::QASYMM8);
+
// Validate interleave kernel
auto_init_if_empty(tmp_a_info, a->clone()->set_tensor_shape(compute_lhs_reshaped_shape(*a, lhs_info, gemm_info.reinterpret_input_as_3d())));
ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMReshapeLHSMatrixKernel::validate(a, &tmp_a_info, lhs_info, gemm_info.reinterpret_input_as_3d()));
@@ -319,12 +295,22 @@
{
TensorInfo mm_result_s32_info{};
- // Output tensor auto inizialitation if not yet initialized
- auto_init_if_empty(mm_result_s32_info, a->clone()->set_tensor_shape(compute_mm_shape(*matrix_a_info, *matrix_b_info, reshape_matrices, reshape_info)).set_data_type(DataType::S32));
+ if(reshape_matrices)
+ {
+ // Output tensor auto inizialitation if not yet initialized
+ auto_init_if_empty(mm_result_s32_info, a->clone()->set_tensor_shape(compute_mm_shape(*matrix_a_info, *matrix_b_info, reshape_info)).set_data_type(DataType::S32));
- // Validate matrix multiply
- ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpMatrixMultiplyKernel::validate(matrix_a_info, matrix_b_info, &mm_result_s32_info, reshape_matrices, reshape_info));
+ // Validate matrix multiply
+ ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpMatrixMultiplyReshapedKernel::validate(matrix_a_info, matrix_b_info, &mm_result_s32_info, lhs_info, rhs_info, reshape_info));
+ }
+ else
+ {
+ // Output tensor auto inizialitation if not yet initialized
+ auto_init_if_empty(mm_result_s32_info, a->clone()->set_tensor_shape(compute_mm_shape(*matrix_a_info, *matrix_b_info, false, reshape_info)).set_data_type(DataType::S32));
+ // Validate matrix multiply
+ ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpMatrixMultiplyKernel::validate(matrix_a_info, matrix_b_info, &mm_result_s32_info, false, reshape_info));
+ }
// Validate offset contribution kernel
ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpOffsetContributionOutputStageKernel::validate(&mm_result_s32_info,
a_offset == 0 ? nullptr : &info_vector_sum_col,
@@ -336,9 +322,16 @@
}
else
{
- // Validate matrix multiply
- ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpMatrixMultiplyKernel::validate(matrix_a_info, matrix_b_info, output, reshape_matrices, reshape_info));
-
+ if(reshape_matrices)
+ {
+ // Validate matrix multiply
+ ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpMatrixMultiplyReshapedKernel::validate(matrix_a_info, matrix_b_info, output, lhs_info, rhs_info, reshape_info));
+ }
+ else
+ {
+ // Validate matrix multiply
+ ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpMatrixMultiplyKernel::validate(matrix_a_info, matrix_b_info, output, false, reshape_info));
+ }
// Validate offset contribution kernel
ARM_COMPUTE_RETURN_ON_ERROR(CLGEMMLowpOffsetContributionKernel::validate(output,
a_offset == 0 ? nullptr : &info_vector_sum_col,
@@ -356,7 +349,7 @@
_memory_group.acquire();
- if(_is_interleaved_transposed)
+ if(_is_gemm_reshaped)
{
// Run reshape matrix A
CLScheduler::get().enqueue(_mtx_a_reshape_kernel, false);
@@ -375,7 +368,14 @@
}
// Run matrix multiply
- CLScheduler::get().enqueue(_mm_kernel, false);
+ if(_is_gemm_reshaped)
+ {
+ CLScheduler::get().enqueue(_mm_reshaped_kernel, false);
+ }
+ else
+ {
+ CLScheduler::get().enqueue(_mm_kernel, false);
+ }
// Run matrix A reduction kernel only if _b_offset is not equal to 0
if(_b_offset != 0)
@@ -401,7 +401,7 @@
{
if(!_is_prepared)
{
- if(_is_interleaved_transposed && _reshape_b_only_on_first_run)
+ if(_is_gemm_reshaped && _reshape_b_only_on_first_run)
{
ARM_COMPUTE_ERROR_ON(!_original_b->is_used());