experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
STT-tensorflow/tensorflow/stream_executor/cuda
History
Mihai Maruseac b550171e78 Prevent CHECK-fail in LSTM/GRU with zero-length input. 
PiperOrigin-RevId: 346239181
Change-Id: I5f233dbc076aab7bb4e31ba24f5abd4eaf99ea4f
2020-12-08 17:41:20 -08:00
..
BUILD
Merge pull request #43237 from benbarsdell:cublaslt
2020-10-15 14:39:38 -07:00
cublas_9_0.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cublas_10_0.inc
Run clang-format on cuda 10.0 inc files, so we can see better diffs for future
2019-08-21 17:51:09 -07:00
cublas_10_1.inc
Add 10.1 inc files for cuda libraries.
2019-08-22 10:25:49 -07:00
cublas_10_2.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cublas_11_0.inc
Add CUDA 11 stub files.
2020-05-26 23:19:19 -07:00
cublas_stub.cc
Relax stub include version checking.
2020-06-17 07:48:07 -07:00
cublasLt_11_0.inc
Merge pull request #43237 from benbarsdell:cublaslt
2020-10-15 14:39:38 -07:00
cublasLt_stub.cc
Add cublasLt wrappers to stream_executor
2020-09-17 15:01:02 +10:00
cuda_9_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cuda_10_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cuda_10_1.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cuda_10_2.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cuda_11_0.inc
Add CUDA 11 stub files.
2020-05-26 23:19:19 -07:00
cuda_activation.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_blas.cc
Merge pull request #44175 from benbarsdell:cublaslt-large-batch-workaround-proper
2020-10-20 10:42:41 -07:00
cuda_blas.h
Merge pull request #44175 from benbarsdell:cublaslt-large-batch-workaround-proper
2020-10-20 10:42:41 -07:00
cuda_diagnostics.cc
Qualify uses of std::string
2020-03-20 13:19:21 -07:00
cuda_diagnostics.h
Qualify uses of std::string
2020-03-20 13:19:21 -07:00
cuda_dnn.cc
Prevent CHECK-fail in LSTM/GRU with zero-length input.
2020-12-08 17:41:20 -08:00
cuda_dnn.h
Roll forward the original CL.
2020-09-09 20:55:27 -07:00
cuda_driver_test.cc
[SE] Don't assume that the CUDA context has not changed in the outermost ScopedActivationContext.
2020-08-10 15:24:22 -07:00
cuda_driver.cc
[SE] Don't assume that the CUDA context has not changed in the outermost ScopedActivationContext.
2020-08-10 15:24:22 -07:00
cuda_driver.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_event.cc
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_event.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_fft.cc
[StreamExecutor] Workaround the cuFFT bug in CUDA 10.1/10.2/11.
2020-06-30 19:14:59 -07:00
cuda_fft.h
[StreamExecutor] Workaround the cuFFT bug in CUDA 10.1/10.2/11.
2020-06-30 19:14:59 -07:00
cuda_gpu_executor.cc
Remove SharedMemoryConfig since it is not used anywhere.
2020-08-11 20:01:04 -07:00
cuda_gpu_executor.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_helpers.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_kernel.cc
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_kernel.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_platform_id.cc
[StreamExecutor] Rename ::perftools::gputools -> ::stream_executor, part 1.
2018-04-17 14:28:51 -07:00
cuda_platform_id.h
[StreamExecutor] Rename ::perftools::gputools -> ::stream_executor, part 1.
2018-04-17 14:28:51 -07:00
cuda_platform.cc
Qualify uses of std::string
2020-03-20 13:19:21 -07:00
cuda_platform.h
Qualify uses of std::string
2020-03-20 13:19:21 -07:00
cuda_rng.cc
Copying cuBLAS and cuDNN headers into separate directories.
2019-05-11 07:25:42 -07:00
cuda_rng.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_runtime_9_0.inc
Add cuda runtime 9.0 API to dlopen wrapper.
2019-04-25 15:20:10 -07:00
cuda_runtime_10_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cuda_runtime_10_1.inc
Add cuda runtime stub for 10.1.
2019-05-03 12:10:46 -07:00
cuda_runtime_10_2.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cuda_runtime_11_0.inc
Add CUDA 11 stub files.
2020-05-26 23:19:19 -07:00
cuda_stream.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cuda_stub.cc
Relax stub include version checking.
2020-06-17 07:48:07 -07:00
cuda_timer.h
PR #25011: [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued)
2019-01-28 11:30:18 -08:00
cudart_stub.cc
Relax stub include version checking.
2020-06-17 07:48:07 -07:00
cudnn_6_0.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cudnn_7_0.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cudnn_7_1.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cudnn_7_3.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cudnn_7_4.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cudnn_7_6.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cudnn_8_0.inc
Eager loading of CUDNN sub-libraries.
2020-10-01 19:51:30 -05:00
cudnn_stub.cc
Roll PR #39577 (CUDNN v8 support) forward with fix:
2020-06-02 12:54:05 -07:00
cudnn_version_test.cc
Clean up cuDNN compatible version check.
2020-07-01 11:26:32 -07:00
cudnn_version.cc
Clean up cuDNN compatible version check.
2020-07-01 11:26:32 -07:00
cudnn_version.h
[SE] Remove some uses of TF string utils in StreamExecutor.
2019-04-26 22:00:34 -07:00
cufft_9_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cufft_10_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cufft_stub.cc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cupti_10_0.inc
Run clang-format on cuda 10.0 inc files, so we can see better diffs for future
2019-08-21 17:51:09 -07:00
cupti_stub.cc
fix build due to cupti static link.
2020-05-29 17:43:42 -07:00
curand_10_0.inc
Run clang-format on cuda 10.0 inc files, so we can see better diffs for future
2019-08-21 17:51:09 -07:00
curand_stub.cc
Copying cuBLAS and cuDNN headers into separate directories.
2019-05-11 07:25:42 -07:00
cusolver_dense_9_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cusolver_dense_10_0.inc
Remove kernels' explicit dependency on cusolver and cusparse.
2019-04-16 15:53:13 -07:00
cusolver_dense_10_1.inc
Add 10.1 inc files for cuda libraries.
2019-08-22 10:25:49 -07:00
cusolver_dense_10_2.inc
Remove redundant cusolverDnIRSInfosGetNiters
2020-03-23 16:04:05 -07:00
cusolver_dense_11_0.inc
Add CUDA 11 stub files.
2020-05-26 23:19:19 -07:00
cusolver_stub.cc
Relax stub include version checking.
2020-06-17 07:48:07 -07:00
cusparse_9_0.inc
Regenerated wrapper includes for all CUDA versions & libraries.
2020-03-19 13:30:00 -07:00
cusparse_10_0.inc
Remove kernels' explicit dependency on cusolver and cusparse.
2019-04-16 15:53:13 -07:00
cusparse_10_1.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cusparse_10_2.inc
Format generated CUDA stub files.
2020-05-26 23:10:31 -07:00
cusparse_11_0.inc
Add CUDA 11 stub files.
2020-05-26 23:19:19 -07:00
cusparse_stub.cc
Relax stub include version checking.
2020-06-17 07:48:07 -07:00
memcpy_test.cc
Remove device memory check, since it's incorrect when the pointer is pointing to pinned host memory. Also, memcpy would fail if the pointer is invalid, so we don't need an additional check.
2019-11-22 14:06:59 -08:00
redzone_allocator_test.cc
Extract GpuAsmOpts struct into its own header file.
2020-02-07 02:18:09 -08:00