27 Commits (13ca364ceba962bbd7eb61bcd2a5dce265678ab4)

Author	SHA1	Message	Date
Huihuang Zheng	12542320c5	Replace TemporaryAllocator by CUDADeviceContextAllocator (#18989) ... 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	6 years ago
Krzysztof Binias	c2c876f718	Fix memory leak in test (#18622) ... * Fix memory leak in test test=develop * Fix memory leak in test test=develop * Fix memory leak in test test=develop * Pull out vars of the loops test=develop	6 years ago
zhhsplendid	124f1df481	Add flags for init and re-alloc gpu ... test=develop	6 years ago
zhhsplendid	22715487dc	add allocator flags ... test=develop	6 years ago
gongweibao	7ab4af2716	Fix brpc compilation. (#15417)	6 years ago
gongweibao	ce70229ba6	Add max_body_size flags to brpc (#15084)	6 years ago
Jiabin Yang	13bc7619f5	Merge pull request #14552 from JiabinYang/fix_mac/fix_pinned_memory ... fix Mac unittest error on reading pined memory flag	6 years ago
JiabinYang	5cd2fc9fd0	just for test	6 years ago
sabreshao	61c5f13fcf	Fix cmake for AMDGPU platform (#13801) ... * HIP cmake. Enable whole archieve build for pybind library. Disable two warning. Rollback to C++11. Link RCCL to WA gpu kernel loading issue. Update eigen to fix build failure. Add more include directories. Fix O3 build failure. Update eigen. fix tensor_util_test segment fault issue add more macro check in hip.cmake. we may consider refine hip.cmake to inherit all add_definitions() in parrent scope, in the future. Fix rocRAND load. Update eigen to fix gru_unit_op and reduce_op. Add HIP support to testing. Update eigen to support int16 and int8 in arg min and arg max. * add rocprim as cub library used by nv implementation * Reduce build time in rocprim. * Add rocprim introduction, remove useless cmake code. * Remove useless flags and format cmake file.	6 years ago
Yu Yang	02631965c8	Refine	6 years ago
Yu Yang	b59a9bfb7c	Clean buffered_allocator ... test=develop	6 years ago
Yu Yang	58ed412f68	refactor(memory): rewrite memory allocation and make it extentable ... Use OO style to rewrite memory allocation.	6 years ago
dzhwinter	4ed0b62476	Move fluid::framework::InitDevices into fluid::platform (#11757) ... * move to platform * "move init from framework to platform" * "remove used init" * "fix ci" * "fix ci" * "fix generic" * "fix ci" * "fix ci" * "fix ci" * "disable fragile test"	7 years ago
tensor-tang	a0c5fd83b2	enable setting initial memory from env	7 years ago
tensor-tang	a427e76958	skip use_mkldnn if do not use it	7 years ago
tensor-tang	6a494380e8	remove mkldnn flag from gtest strdup	7 years ago
Luo Tao	c6d230e03e	add FLAGS_use_mkldnn to global control use_mkldnn	7 years ago
Xin Pan	b138d29c38	Avoid init_p2p all the times	7 years ago
qijun	d2e7276377	fix warp ctc dir error	7 years ago
Yi Wang	fc374821dd	Correct #include path	7 years ago
Yu Yang	ef1aba39a6	Rewrite mixed_vector.h	7 years ago
Luo Tao	d2caf777ae	set FLAGS_warpctc_dir to pass the test_warpctc_op unit test	7 years ago
Liu Yiqun	6ac2e079b3	Enable whole-archive flag in cc_test and use cc_test to rewrite the CMakeLists.txt of inference unittest.	7 years ago
dzhwinter	5f98500009	Make init device on all gpu by default (#7345) ... * "init use all default devices" * "fix init test"	7 years ago
dzhwinter	0d2235aadf	GPUPlace to CUDAPlace (#6960)	7 years ago
dzhwinter	735eba2976	Feature/operator run place (#6783) ... * "change operator interface" * "move devicepool to device_context" * "fix operator test" * "fix op_registry Run interface" * "net op passed. Need to fix nccl multi-Context" * "add nccl group function" * "add nccl group function" * "fix gpu count exceed 32 error" * "fix recurrent op, nccl op" * "change the other operators interface with Place" * "fix typo" * "fix pybind" * "fix device in python side" * "fix pybind failed" * "add init for test" * "fix CI"	7 years ago
QI JUN	1238706d72	Refine unittest with setting gflags (#5476) ... * add gflags for C++ unittest	7 years ago