/aosp_15_r20/external/tensorflow/tensorflow/compiler/xla/stream_executor/cuda/ |
H A D | cuda_fft.cc | 100 fft::Type type, int batch_count, ScratchAllocator *scratch_allocator) { in Initialize() 259 ScratchAllocator *scratch_allocator) { in Initialize() 268 Stream *stream, ScratchAllocator *scratch_allocator) { in UpdateScratchAllocator() 332 ScratchAllocator *scratch_allocator) { in Create1dPlanWithScratchAllocator() 365 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create2dPlanWithScratchAllocator() 400 bool in_place_fft, ScratchAllocator *scratch_allocator) { in Create3dPlanWithScratchAllocator() 448 bool in_place_fft, int batch_count, ScratchAllocator *scratch_allocator) { in CreateBatchedPlanWithScratchAllocator() 473 Stream *stream, fft::Plan *plan, ScratchAllocator *scratch_allocator) { in UpdatePlanWithScratchAllocator()
|
H A D | cuda_blas.cc | 1193 ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 1330 ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 1347 ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 1363 ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 1380 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl() 1397 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasInternalImpl()
|
H A D | cuda_dnn.cc | 2892 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionForwardWorkspace() 2941 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionBackwardDataWorkspace() 2991 ScratchAllocator* scratch_allocator) { in AllocateCudnnConvolutionBackwardFilterWorkspace() 3058 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionForwardAlgorithm() 3121 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionBackwardDataAlgorithm() 3182 const CudnnTensorDescriptor& output_nd, ScratchAllocator* scratch_allocator, in GetCudnnConvolutionBackwardFilterAlgorithm() 3964 ScratchAllocator* scratch_allocator, dnn::AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 5765 ScratchAllocator* scratch_allocator, in DoFusedConvolve() 5843 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss()
|
H A D | cuda_blas_lt.cc | 262 ScratchAllocator& scratch_allocator, in DoMatmul()
|
/aosp_15_r20/external/tensorflow/tensorflow/stream_executor/rocm/ |
H A 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()
|
H A D | rocm_blas.cc | 712 int batch_count, uint64_t batch_stride, ScratchAllocator *scratch_allocator, in AllocateStridedBuffer() 771 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatchedInternal() 865 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 887 int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 905 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 925 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched() 945 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in DoBlasGemmBatched()
|
H A D | rocm_dnn.cc | 2451 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss() 2911 ScratchAllocator* scratch_allocator, dnn::AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 3154 ScratchAllocator* scratch_allocator, in GetConvolveRunners() 3219 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithms() 3241 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithmsImmediateMode() 3450 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithmsFindMode() 3807 ScratchAllocator* scratch_allocator, in DoFusedConvolve()
|
/aosp_15_r20/external/tensorflow/tensorflow/core/kernels/ |
H A D | matmul_op_fused.cc | 204 BlasScratchAllocator scratch_allocator(context); in AutotuneMatmul() local 304 se::blas::ProfileResult* profile_result) { in operator ()() 318 BlasScratchAllocator scratch_allocator(context); in operator ()() local
|
H A D | conv_ops_gpu.h | 133 AllocateScratchOrFallback(se::ScratchAllocator* scratch_allocator, in AllocateScratchOrFallback() 162 DnnScratchAllocator* scratch_allocator, in LaunchAutotunedConv()
|
H A D | cudnn_pooling_gpu.cc | 108 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local 244 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local
|
H A D | conv_ops_gpu.cc | 55 DnnScratchAllocator scratch_allocator(scratch_size_limit, ctx); in AutotuneConvImpl() local 384 DnnScratchAllocator scratch_allocator(scratch_size_limit, ctx); in AutotuneUnfusedConv() local
|
H A D | pooling_ops_common.cc | 398 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local 722 DnnScratchAllocator scratch_allocator(PoolingScratchSize, context); in Compute() local
|
H A D | conv_grad_input_ops.cc | 386 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, ctx); in operator ()() local
|
H A D | conv_ops_3d.cc | 513 DnnScratchAllocator scratch_allocator(ConvolveScratchSize, ctx); in launch() local
|
H A D | conv_grad_ops_3d.cc | 1529 DnnScratchAllocator scratch_allocator(ConvolveBackwardDataScratchSize, in Compute() local 1928 DnnScratchAllocator scratch_allocator(ConvolveBackwardFilterScratchSize, in Compute() local
|
H A D | fft_ops.cc | 479 CufftScratchAllocator scratch_allocator(CufftScratchSize, ctx); in DoFFT() local
|
/aosp_15_r20/external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
H A D | gpu_conv_algorithm_picker.cc | 190 ScratchAllocator* scratch_allocator, se::Stream* stream) { in GetMIOpenAlgorithms() 489 se::RedzoneAllocator scratch_allocator( in AutotuneOneConvRunner() local 828 ScratchAllocator scratch_allocator(device_ordinal, allocator); in PickBestAlgorithmNoCacheRocm() local
|
H A D | matmul_utils.cc | 693 se::ScratchAllocator& scratch_allocator, in DoMatmul() 722 se::ScratchAllocator& scratch_allocator, in ExecuteOnStream()
|
H A D | cublas_lt_matmul_thunk.cc | 61 se::OwningScratchAllocator<> scratch_allocator(allocs.device_ordinal(), in ExecuteOnStream() local
|
H A D | fft_thunk.cc | 104 se::OwningScratchAllocator<2> scratch_allocator(device_ordinal, in RunFft() local
|
H A D | gemm_algorithm_picker.cc | 296 se::OwningScratchAllocator<> scratch_allocator( in DoGemmAutotune() local
|
/aosp_15_r20/external/tensorflow/tensorflow/compiler/xla/stream_executor/ |
H A D | stream.h | 344 ScratchAllocator *scratch_allocator, in ConvolveWithAlgorithm() 377 DeviceMemory<OutputT> *output, ScratchAllocator *scratch_allocator, in FusedConvolveWithAlgorithm()
|
H A D | dnn.h | 1312 DeviceMemoryBase output_data, ScratchAllocator* scratch_allocator, in DoFusedConvolve() 1330 ScratchAllocator* scratch_allocator, AlgorithmDesc* algorithm_desc, in PrepareForConvolution() 2626 ScratchAllocator* scratch_allocator, AlgorithmDesc* algorithm_desc, in DoPrepareForConvolution() 2640 ScratchAllocator* scratch_allocator, DeviceMemory<uint8>* scratch_memory, in DoPrepareForCtcLoss()
|
H A D | stream_executor_pimpl.cc | 301 ScratchAllocator* scratch_allocator, in GetConvolveRunners() 344 ScratchAllocator* scratch_allocator, in GetMIOpenConvolveAlgorithms()
|
H A D | stream.cc | 1716 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 1755 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 1792 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 1838 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch() 1874 int ldc, int batch_count, ScratchAllocator *scratch_allocator) { in ThenBlasGemmBatchedWithScratch()
|