experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
master
STT-tensorflow/tensorflow/stream_executor/gpu
History
A. Unique TensorFlower c7e57df256 Use device handle instead of gpu ordinal in GpuVirtualMemAllocator for 
configuring peer access.

Check that peers support virtual address management and have peer access
at GpuVirtualMemAllocator creation time.

PiperOrigin-RevId: 353124035
Change-Id: I8549253b84af35a664e9a6810c6b40deba20a532
2021-01-21 16:11:33 -08:00
..
asm_compiler.cc Be friendlier to older ptxas versions 2020-12-23 16:17:52 -08:00
asm_compiler.h [XLA/GPU] Add parallel compilation for LLVM and ptxas. 2020-12-08 09:22:49 -08:00
BUILD Be friendlier to older ptxas versions 2020-12-23 16:17:52 -08:00
gpu_activation.cc PR : [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued) 2019-01-28 11:30:18 -08:00
gpu_activation.h PR : [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued) 2019-01-28 11:30:18 -08:00
gpu_asm_opts.h Add extra_flags to ptxas for more control over the behavior. 2020-03-17 00:21:30 -07:00
gpu_diagnostics.h Qualify uses of std::string 2020-03-20 13:19:21 -07:00
gpu_driver.h Use device handle instead of gpu ordinal in GpuVirtualMemAllocator for 2021-01-21 16:11:33 -08:00
gpu_event.cc [SE] Rename se::gpu::GpuDriver::CreateEvent so it does not conflict with a windows.h macro. 2019-05-02 18:04:22 -07:00
gpu_event.h
gpu_executor.h Remove SharedMemoryConfig since it is not used anywhere. 2020-08-11 20:01:04 -07:00
gpu_helpers.h Use float2/double2 to pass complex float/double values to cublas. 2020-04-15 18:04:26 -07:00
gpu_kernel.h PR : [GPU][ROCm][CUDA] StreamExecutor logic for ROCm / CUDA platform (PR 20709 / 22669 / 24156 continued) 2019-01-28 11:30:18 -08:00
gpu_rng.h Qualify uses of std::string 2020-03-20 13:19:21 -07:00
gpu_stream.cc Implement GPU device priority for virtual devices. 2020-06-02 22:31:35 -07:00
gpu_stream.h Implement GPU device priority for virtual devices. 2020-06-02 22:31:35 -07:00
gpu_timer.cc [SE] Rename se::gpu::GpuDriver::CreateEvent so it does not conflict with a windows.h macro. 2019-05-02 18:04:22 -07:00
gpu_timer.h minor spelling tweaks 2020-01-16 14:36:52 +09:00
gpu_types.h Copying cuBLAS and cuDNN headers into separate directories. 2019-05-11 07:25:42 -07:00
redzone_allocator.cc Extract GpuAsmOpts struct into its own header file. 2020-02-07 02:18:09 -08:00
redzone_allocator.h Add a clarifying comment; NFC 2020-07-28 12:52:09 -07:00