- 01 9月, 2022 1 次提交
-
-
由 Leo Chen 提交于
* refine cmake of framework * add deps for dense tensor * fix deps * remove alloc(ctx) * add depends on mkldnn
-
- 01 8月, 2022 1 次提交
-
-
由 Leo Chen 提交于
* remove cudaDeviceContext * remove more template * fix rocm compile * remove alias name CUDADeviceContext * fix compile * fix tests * revert changes
-
- 26 6月, 2022 1 次提交
-
-
由 Sing_chan 提交于
-
- 01 4月, 2022 1 次提交
-
-
由 wanghuancoder 提交于
* support pinned, test=develop * support async_write, test=develop * refine, test=develop * refine, test=develop * refine, test=develop * refine,test=develop * refine, test=develop * refine, test=develop * refine, test=develop * refine, test=develop
-
- 27 3月, 2022 1 次提交
-
-
由 From00 提交于
* Make StreamSafeCUDAAllocator compatible with NaiveBestFit strategy * Set FLAGS_use_stream_safe_cuda_allocator to false * Update * Remove unnecessary code * Fix CI errors * Add UT
-
- 06 2月, 2022 1 次提交
-
-
由 Wilber 提交于
-
- 22 2月, 2021 1 次提交
-
-
由 Qi Li 提交于
-
- 19 1月, 2021 1 次提交
-
-
由 Leo Chen 提交于
* unify calling cudaSetDevice * fix compile
-
- 11 9月, 2019 1 次提交
-
-
由 Huihuang Zheng 提交于
TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory. We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton. Also added data_feed_proto to operator to fix CI in CPU compilation
-