| /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() 162 se::ScratchAllocator* scratch_allocator, in RunGpuConvInternalImpl() 210 se::ScratchAllocator* scratch_allocator, in RunGpuConvInternalImpl() 234 se::ScratchAllocator* scratch_allocator, in RunGpuConvImpl() 536 ScratchBufAllocator scratch_allocator(scratch_buf); in RunGpuConv() local 544 se::ScratchAllocator* scratch_allocator, se::Stream* stream, in RunGpuConv()
|
| D | gpu_conv_algorithm_picker.cc | 129 ScratchAllocator* scratch_allocator, se::Stream* stream) { in GetMIOpenAlgorithms() 437 se::RedzoneAllocator scratch_allocator( in PickBestAlgorithmNoCacheCuda() local 629 ScratchAllocator scratch_allocator(device_ordinal, allocator); in PickBestAlgorithmNoCacheRocm() local
|
| D | fft_thunk.cc | 125 FftScratchAllocator scratch_allocator(buffer_allocations.device_ordinal(), in ExecuteOnStream() local
|
| /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 | 2832 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionForwardWorkspace() 2876 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionBackwardDataWorkspace() 2922 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionBackwardFilterWorkspace() 2985 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionForwardAlgorithm() 3048 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionBackwardDataAlgorithm() 3109 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionBackwardFilterAlgorithm() 3784 ScratchAllocator* scratch_allocator) { in GetFirstWorkingExecutionPlan() 3871 ScratchAllocator* scratch_allocator, dnn::AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 4068 ScratchAllocator* scratch_allocator, in DoConvolveWithExecutionPlan() 4879 ScratchAllocator* scratch_allocator, in DoFusedConvolve() [all …]
|
| D | cuda_blas.cc | 2314 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2450 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2468 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2484 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2502 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 2520 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 3179 ScratchAllocator *scratch_allocator, in DoBlasInternalImpl() 3481 DeviceMemoryBase c, DeviceMemoryBase d, ScratchAllocator *scratch_allocator, in DoBlasInternalImpl() 3643 DeviceMemoryBase c, ScratchAllocator *scratch_allocator, in DoBlasInternalImpl()
|
| /external/tensorflow/tensorflow/stream_executor/rocm/ |
| D | rocm_fft.cc | 161 fft::Type type, int batch_count, ScratchAllocator *scratch_allocator) { in Initialize() 322 ScratchAllocator *scratch_allocator) { in Initialize() 331 Stream *stream, ScratchAllocator *scratch_allocator) { in UpdateScratchAllocator() 391 ScratchAllocator *scratch_allocator) { in Create1dPlanWithScratchAllocator() 420 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create2dPlanWithScratchAllocator() 450 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create3dPlanWithScratchAllocator() 485 bool in_place_fft, int batch_count, ScratchAllocator *scratch_allocator) { in CreateBatchedPlanWithScratchAllocator() 500 Stream *stream, fft::Plan *plan, ScratchAllocator *scratch_allocator) { in UpdatePlanWithScratchAllocator()
|
| D | rocm_blas.cc | 1671 int batch_count, uint64_t batch_stride, ScratchAllocator *scratch_allocator, in AllocateStridedBuffer() 1730 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatchedInternal() 1824 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 1846 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 1864 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 1884 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 1904 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 2363 DeviceMemoryBase c, ScratchAllocator *scratch_allocator, in DoBlasLtMatmul()
|
| D | rocm_dnn.cc | 2448 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss() 2908 ScratchAllocator* scratch_allocator, dnn::AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 3098 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithms() 3120 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithmsImmediateMode() 3329 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithmsFindMode() 3682 ScratchAllocator* scratch_allocator, in DoFusedConvolve()
|
| /external/tensorflow/tensorflow/stream_executor/ |
| D | stream.h | 338 DeviceMemory<OutputType> *output, ScratchAllocator *scratch_allocator, in ConvolveWithAlgorithm() 367 DeviceMemory<OutputType> *output, ScratchAllocator *scratch_allocator, in ConvolveWithExecutionPlan() 397 DeviceMemory<OutputT> *output, ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm() 425 DeviceMemory<OutputT> *output, ScratchAllocator *scratch_allocator, in FusedConvolveWithExecutionPlan() 462 ScratchAllocator *scratch_allocator, in ConvolveBackwardDataWithExecutionPlan() 490 ScratchAllocator *scratch_allocator, in ConvolveBackwardDataWithAlgorithm() 522 ScratchAllocator *scratch_allocator, in ConvolveBackwardFilterWithAlgorithm() 554 ScratchAllocator *scratch_allocator, in ConvolveBackwardFilterWithExecutionPlan()
|
| D | dnn.h | 1264 DeviceMemoryBase output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 1282 ScratchAllocator* scratch_allocator, AlgorithmDesc* algorithm_desc, in PrepareForConvolution() 2594 ScratchAllocator* scratch_allocator, AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 2608 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss()
|
| D | dnn.cc | 76 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithms()
|
| D | stream_executor_pimpl.cc | 297 ScratchAllocator *scratch_allocator, in GetMIOpenConvolveAlgorithms()
|
| D | stream.cc | 3610 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 3642 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 3674 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 3710 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 3749 int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 3773 ScratchAllocator *scratch_allocator, in ThenBlasLtMatmulImpl()
|
| /external/tensorflow/tensorflow/core/kernels/ |
| D | conv_ops_3d.cc | 554 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in launch() local 611 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in launch() local 678 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in launch() local
|
| D | conv_grad_input_ops.cc | 356 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, ctx); in operator ()() local 436 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in operator ()() local 494 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, ctx); in operator ()() local
|
| 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_ops_3d.cc | 1562 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in Compute() local 1622 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in Compute() local 1670 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in Compute() local 2114 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in Compute() local 2160 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in Compute() local 2225 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in Compute() local
|
| D | pooling_ops_common.cc | 398 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local 716 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local
|
| D | conv_grad_filter_ops.cc | 1041 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in operator ()() local 1100 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in operator ()() local 1169 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, ctx); in operator ()() local
|
| D | conv_ops.cc | 1164 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in operator ()() local 1220 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in operator ()() local 1289 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in operator ()() local
|
| D | fft_ops.cc | 475 CufftScratchAllocator scratch_allocator(CufftScratchSize, ctx); in DoFFT() local
|