/external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
D | gpu_conv_runner.cc | 77 se::ScratchAllocator* scratch_allocator, in RunGpuConvForward() 97 se::ScratchAllocator* scratch_allocator, in RunGpuConvForwardActivation() 150 se::ScratchAllocator* scratch_allocator, in RunGpuConvInternalImpl() 198 se::ScratchAllocator* scratch_allocator, in RunGpuConvInternalImpl() 222 se::ScratchAllocator* scratch_allocator, in RunGpuConvImpl() 500 ScratchBufAllocator scratch_allocator(scratch_buf); in RunGpuConv() local 508 se::ScratchAllocator* scratch_allocator, se::Stream* stream, in RunGpuConv()
|
D | gpu_conv_algorithm_picker.cc | 128 ScratchAllocator* scratch_allocator, se::Stream* stream) { in GetMIOpenAlgorithms() 463 se::RedzoneAllocator scratch_allocator( in PickBestAlgorithmNoCacheCuda() local 689 ScratchAllocator scratch_allocator(device_ordinal, allocator); in PickBestAlgorithmNoCacheRocm() local
|
D | fft_thunk.cc | 126 FftScratchAllocator scratch_allocator(buffer_allocations.device_ordinal(), in ExecuteOnStream() local
|
/external/tensorflow/tensorflow/stream_executor/rocm/ |
D | rocm_fft.cc | 161 fft::Type type, int batch_count, ScratchAllocator *scratch_allocator) { in Initialize() 321 ScratchAllocator *scratch_allocator) { in Initialize() 330 Stream *stream, ScratchAllocator *scratch_allocator) { in UpdateScratchAllocator() 389 ScratchAllocator *scratch_allocator) { in Create1dPlanWithScratchAllocator() 418 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create2dPlanWithScratchAllocator() 448 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create3dPlanWithScratchAllocator() 483 bool in_place_fft, int batch_count, ScratchAllocator *scratch_allocator) { in CreateBatchedPlanWithScratchAllocator() 498 Stream *stream, fft::Plan *plan, ScratchAllocator *scratch_allocator) { in UpdatePlanWithScratchAllocator()
|
D | rocm_dnn.cc | 2397 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss() 2851 ScratchAllocator* scratch_allocator, dnn::AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 3104 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithms() 3126 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithmsImmediateMode() 3335 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithmsFindMode() 3685 DeviceMemory<double>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 3701 DeviceMemory<float>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 3718 DeviceMemory<Eigen::half>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 3734 DeviceMemory<int8>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve()
|
D | rocm_blas.cc | 1986 int batch_count, uint64_t batch_stride, ScratchAllocator *scratch_allocator, in AllocateStridedBuffer() 2045 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatchedInternal() 2139 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 2161 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 2179 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 2199 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 2219 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 2676 DeviceMemoryBase c, ScratchAllocator *scratch_allocator, in DoBlasLtMatmul()
|
/external/tensorflow/tensorflow/stream_executor/cuda/ |
D | cuda_fft.cc | 81 fft::Type type, int batch_count, ScratchAllocator *scratch_allocator) { in Initialize() 238 ScratchAllocator *scratch_allocator) { in Initialize() 247 Stream *stream, ScratchAllocator *scratch_allocator) { in UpdateScratchAllocator() 313 ScratchAllocator *scratch_allocator) { in Create1dPlanWithScratchAllocator() 344 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create2dPlanWithScratchAllocator() 377 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create3dPlanWithScratchAllocator() 423 bool in_place_fft, int batch_count, ScratchAllocator *scratch_allocator) { in CreateBatchedPlanWithScratchAllocator() 447 Stream *stream, fft::Plan *plan, ScratchAllocator *scratch_allocator) { in UpdatePlanWithScratchAllocator()
|
D | cuda_dnn.cc | 2540 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionForwardWorkspace() 2584 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionBackwardDataWorkspace() 2630 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionBackwardFilterWorkspace() 2693 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionForwardAlgorithm() 2756 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionBackwardDataAlgorithm() 2817 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionBackwardFilterAlgorithm() 2991 ScratchAllocator* scratch_allocator, dnn::AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 3188 ScratchAllocator* scratch_allocator, in DoFusedConvolveImpl() 3766 DeviceMemory<double>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 3788 DeviceMemory<float>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() [all …]
|
D | cuda_blas.cc | 2370 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2519 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2537 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2553 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2571 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2589 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 3227 ScratchAllocator *scratch_allocator, in DoBlasInternalImpl() 3529 DeviceMemoryBase c, DeviceMemoryBase d, ScratchAllocator *scratch_allocator, in DoBlasInternalImpl() 3691 DeviceMemoryBase c, ScratchAllocator *scratch_allocator, in DoBlasInternalImpl()
|
/external/tensorflow/tensorflow/stream_executor/ |
D | stream.h | 333 DeviceMemory<OutputType> *output, ScratchAllocator *scratch_allocator, in ConvolveWithAlgorithm() 444 ScratchAllocator *scratch_allocator, in ConvolveBackwardDataWithAlgorithm() 476 ScratchAllocator *scratch_allocator, in ConvolveBackwardFilterWithAlgorithm()
|
D | dnn.h | 1153 DeviceMemory<double>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 1171 DeviceMemory<float>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 1192 ScratchAllocator* scratch_allocator, in DoFusedConvolve() 1211 DeviceMemory<int8>* output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 1251 ScratchAllocator* scratch_allocator, AlgorithmDesc* algorithm_desc, in PrepareForConvolution() 2686 ScratchAllocator* scratch_allocator, AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 2700 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss()
|
D | dnn.cc | 59 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithms()
|
D | stream.cc | 456 ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm() 487 ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm() 519 DeviceMemory<Eigen::half> *output, ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm() 550 ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm() 581 ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm() 4058 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 4090 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 4122 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 4158 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 4197 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() [all …]
|
D | stream_executor_pimpl.cc | 284 ScratchAllocator *scratch_allocator, in GetMIOpenConvolveAlgorithms()
|
/external/tensorflow/tensorflow/core/kernels/ |
D | cudnn_pooling_gpu.cc | 107 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local 249 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local
|
D | conv_grad_input_ops.cc | 359 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, ctx); in operator ()() local 410 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in operator ()() local
|
D | conv_ops_3d.cc | 531 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in launch() local 624 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in launch() local
|
D | pooling_ops_common.cc | 395 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local 713 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local
|
D | conv_grad_ops_3d.cc | 1444 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in Compute() local 1521 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in Compute() local 1935 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in Compute() local 2014 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in Compute() local
|
D | conv_grad_filter_ops.cc | 1004 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in operator ()() local 1099 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, ctx); in operator ()() local
|
D | conv_ops.cc | 1020 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in operator ()() local 1115 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in operator ()() local
|
D | fft_ops.cc | 468 CufftScratchAllocator scratch_allocator(CufftScratchSize, ctx); in DoFFT() local
|