95 Commits (438e261bbc8f00a1bbe43bb92460e2279ba9b685)

Author SHA1 Message Date
  臧庆香 1dd489096e profiler pynative python code 4 years ago
  i-robot 476ec8007b
!30775 Support GPU cluster step trace. 4 years ago
  i-robot e0de84afec
!30787 profiler supports pynative mode 4 years ago
  liuchuting cf2877d86e Support cluster step trace. 4 years ago
  臧庆香 1fdd76c50f profiler pynative cpp 4 years ago
  zhoufeng f49b195c39 extract common as an independent shared library 4 years ago
  fangzehua e64055d2ca add parallel profiling 4 years ago
  marui d47e79b04c Refactor ccsrc directories and CMakeLists files 4 years ago
  liuchuting 2a9a968f7f Profiler memory could not collect data. 4 years ago
  yanghaitao1 c9d2bfad7b report cann error msg if failed to call cann api 4 years ago
  liuchuting 6f6f21126d Fixed an issue where GPU multi-card network iteration tail time was negative. 4 years ago
  liuchuting 0686651e9b Fixed an issue where GPU multi-card network iteration tail time was negative. 4 years ago
  yuximiao 5e494d71e4 report cann msg if cann api return error 4 years ago
  yuximiao 393d3621a3 fix no paralle stratege if start profiler in the process of trainging 4 years ago
  臧庆香 1c82f79320 Add ACL_PROF_HCCL_TRACE Switch 4 years ago
  i-robot 812dba7e21 !24771 add compile option H to visible hidden interface 4 years ago
  yuximiao e99c0a48e6 support start profiler in the minddle of training. 4 years ago
  He Wei 41dcac9c49 Replace std::unordered_map/set with robin-hood-hashing 4 years ago
  ougongchang c01dbf25c9 Change the parallel strategt data to JSON format 4 years ago
  i-robot 8d603ffc41 !26468 Change profiler parallel strategy file mode to 600, keep same with the other files 4 years ago
  i-robot ce00ee1ad1 !25367 use acl api to control profiling 4 years ago
  ougongchang 744e291982 Change profiler parallel strategy file mode to 600, keep same with the 4 years ago
  yanghaitao1 c94aa6b872 use profiler acl api instead 4 years ago
  ougongchang a748d2c8af fix profiling parallel strategy can not get data in SaveCompileGraph phase 4 years ago
  liuchuting bb5bf3f289 repair warning of memcpy,memset 4 years ago
  ougongchang 9229f1c1ff profiler support to collect parallel strategy info 4 years ago
  yelihua cfa8c7a0e8 update OWNERS 4 years ago
  healing 56ae399228 open gpu profiling interface 4 years ago
  yanghaitao1 a30b940ad3 add commiter to OWNERS files 4 years ago
  i-robot 9f92ac51b1 !24250 fix some review issues in profiler 4 years ago
  yanghaitao1 09e2d4bb88 fix some review issues in profiler 4 years ago
  yanghaitao1 6a92b2c538 fix reviews issues in profiler 4 years ago
  i-robot a009c8e8b4 !23542 fix pclint in profiler 4 years ago
  yanghaitao1 56fd6f57e7 fix pclint in profiler 4 years ago
  yanghaitao1 177f3f75bf remove profiler if compiled with -s on 4 years ago
  i-robot 94f7cba97c !23219 remove profiling parameters in set_context function 4 years ago
  zhangyihui e121fb2e8c Heterogeneous cpu performance in GPU 4 years ago
  yanghaitao1 83302ad23c remove profiling parameters in set_context function 4 years ago
  i-robot 18078a6457 !22917 If users need GPU profiling, they need to use the root account to start the process 4 years ago
  臧庆香 2bc81c48e0 profiling GPU permission 4 years ago
  zhangyihui 6a36171a0e transform device_id to rank_id for cpu_profiler 4 years ago
  zhangyihui 3d19949eb4 device_id to rank_id 4 years ago
  ms_yan 36a8886ca2 Revert "[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset" 4 years ago
  djc b077aa1cab [feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset 4 years ago
  djc 4e6f7dc97d [feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset 4 years ago
  lvchangquan e8d9803258 add profiling codes based on ascend and gpu in pynative mode 4 years ago
  gaojing 310841bd51 profiler cleancode 4 years ago
  gzhcv 96dfea1e6f Add scope-level flops in ascend profiler 4 years ago
  gzhcv 4c78a47184 Fit Gpu LoopCount for profiler module v2 4 years ago
  limingqi107 406c252834 enable mindRT 4 years ago