Commit Graph

512 Commits

Author SHA1 Message Date
i-robot d9b8429631 !19424 Fixing CPU dump issue
Merge pull request !19424 from parastooashtari/cpu_dump
2021-07-06 03:58:32 +00:00
i-robot 7dffa5096c !19427 Fix offline dbg issue: multiple read_tensor output if dumping twice
Merge pull request !19427 from TinaMengtingZhang/multi_output_when_dump_twice
2021-07-06 03:15:12 +00:00
i-robot 2d28931533 !19114 [MS][RDR] Add method for checking GLOG_log_dir in C++ module
Merge pull request !19114 from louie5/master
2021-07-06 01:07:12 +00:00
TinaMengtingZhang 4147e1bd9a get multiple dump result when dump twice 2021-07-05 17:00:02 -04:00
Parastoo Ashtari a86a38080f Fixed CPU dump error 2021-07-05 10:37:05 -04:00
louie5 ef98f259ce create directory of log files using 'GLOG_log_dir' in c++ frontend while it does not exist
check the real path of directory whether valid
2021-07-05 15:21:12 +08:00
i-robot fa0d79478d !19169 actor runtime support CPU dump
Merge pull request !19169 from limingqi107/actor_runtime2
2021-07-04 14:48:52 +00:00
i-robot 66c03ebc06 !19314 Fix receive WaitCMD instead of terminate for multigraph on GPU issue
Merge pull request !19314 from parastooashtari/master
2021-07-04 03:20:45 +00:00
limingqi107 e761655a42 actor runtime support CPU dump 2021-07-04 11:19:09 +08:00
Parastoo Ashtari a153f6b35f Fix receive WaitCMD instead of terminate for multigraph GPU 2021-07-02 14:30:00 -04:00
zhousiyi a5b4e7bbf8 lift fv before grad except weight, then convert
switch(cond, partial(g1, xs), partial(g2, ys))(Zs)
 to
 switch(cond, g1, g2)(Xs, Ys, Zs)

 switch_layer(index, make_tuple(partial(g1, xs), partial(g2, ys)))(Zs)
 to
 switch_layer(index, make_tuple(g1, g2))(Xs, Ys, Zs)

put Zs at last when unifyparameter as it may have u-monad or io-monad

use joined args other than broadened one as some extra parameter which is not a parameter of while_header can be add to while_body

inline fprop_switch forcely

reorder the parameter if one of the parameter is Monad when incorporate call

incorporate switch tuple_getitem if item 0 of tuple is EnvInstance or
item 1 of tuple is bprop function

addn with shape() and shape(1)

remove context_ from FuncGraphEvaluator to make it re-entry able to resolve evaluator stuck issue because of re-entry of the same FuncGraphEvaluator
2021-07-03 00:03:37 +08:00
i-robot 35e35d5af4 !19077 Trace CNode Enter&Leave during StackFrame jump.
Merge pull request !19077 from 张清华/opt
2021-07-01 03:07:56 +00:00
i-robot a230e0e62c !19185 Fix missing parameter name issue in dump and incorrect read tensor output size in offline debugger
Merge pull request !19185 from TinaMengtingZhang/bugfix_miss_param_name
2021-07-01 02:14:03 +00:00
i-robot bb76628b70 !18983 infer optimize to support determined fun first goes
Merge pull request !18983 from lanzhineng/infer_optv3
2021-07-01 02:00:02 +00:00
TinaMengtingZhang 9a68c26fc6 Fix issues: 1.missing parameter name in dump; 2.too much read tensor output in offline debugger 2021-06-30 10:29:40 -04:00
Parastoo Ashtari 4e4be17648 SetDebugger for MindRTBackend and clean debug_actor code 2021-06-29 12:25:11 -04:00
limingqi107 406c252834 enable mindRT 2021-06-29 17:44:05 +08:00
Zhang Qinghua 3b85a38c9d Trace CNode Enter&Leave during StackFrame jump. 2021-06-29 14:31:48 +08:00
lanzhineng e2adafed3b infer opt: determined fun graph first runs & opt exception handle 2021-06-29 10:26:05 +08:00
Parastoo Ashtari e2a4172560 make gpu dump truly async 2021-06-28 16:44:35 -04:00
i-robot 3629df552c !18937 upgrade_ascend_0626_mindspore
Merge pull request !18937 from mindspore_ding/upgrade_ascend_0626
2021-06-28 15:50:14 +00:00
dingpeifei 63784e49f5 upgrade_ascend_0626_mindspore 2021-06-28 22:03:19 +08:00
i-robot 97fa0e6398 !18613 fix issues for code check
Merge pull request !18613 from jiangshuqiang/fix_code_check
2021-06-28 12:22:55 +00:00
i-robot 007707904e !18897 use -1 iteration for init state and fix missing tensor read outputs
Merge pull request !18897 from john_tzanakakis/jt_bug_fixes
2021-06-26 02:53:08 +00:00
i-robot 9886aa613e !18845 Fix dump failure when const node is string type
Merge pull request !18845 from TinaMengtingZhang/bugfix_dump_fail_with_summaryop
2021-06-26 02:23:06 +00:00
John Tzanakakis c85d65c0ef use -1 iteration for init state and fix missing tensor read outputs 2021-06-25 19:47:54 -04:00
i-robot 29e42efd98 !18867 Delete unnecessary logs in dump
Merge pull request !18867 from maning202007/delete_unnecessary_log
2021-06-25 13:09:48 +00:00
i-robot c986916d48 !18675 [Debugger] Update two arguments "is_output" and "rank_id" in offline debugger python API
Merge pull request !18675 from TinaMengtingZhang/new_arg_is_output
2021-06-25 10:33:46 +00:00
maning202007 cb6e21a16a Delete unnecessary logs in dump 2021-06-25 15:06:32 +08:00
jiangshuqiang 0c9d2629b1 fix code check issues 2021-06-25 10:26:01 +08:00
TinaMengtingZhang 3ec5c760d7 Fix dump failed with StringImm node (i.e. summary op) 2021-06-24 12:28:51 -04:00
i-robot 708e56f659 !17899 Fix compile cache bug for resent50
Merge pull request !17899 from LiangZhibo/cache
2021-06-24 13:29:58 +00:00
i-robot d621cdbc97 !18790 validate dump iteration and net_name, use device shape for trans_flag false
Merge pull request !18790 from john_tzanakakis/jt_bug_fixes
2021-06-24 08:08:11 +00:00
l00591931 8ae5d7cc84 Fix compile cache for resnet50 2021-06-24 14:36:24 +08:00
i-robot 6d254ac409 !18679 call nnacl implement for matmul and batchmatmul on ARM
Merge pull request !18679 from zhangbuxue/call_nnacl_implement_for_matmul_and_batchmatmul_on_ARM
2021-06-24 06:14:22 +00:00
buxue f3dea76b25 call nnacl implement for matmul and batchmatmul on ARM 2021-06-24 09:29:45 +08:00
Parastoo Ashtari 8d248cd925 Add debug info for online debugger 2021-06-23 19:00:09 -04:00
John Tzanakakis baad7b48d3 validate dump iteration and net_name, use device shape for trans_flag false 2021-06-23 17:46:49 -04:00
TinaMengtingZhang 4e8db5e115 change arg 'is_parameter' to 'is_output' and rename device_id to rank_id in python api 2021-06-23 14:25:53 -04:00
i-robot d4aca69981 !18448 Add debugger to new GPU runtime
Merge pull request !18448 from parastooashtari/new_unified_gpu
2021-06-23 01:56:09 +00:00
i-robot c2e03304ef !18695 [Bug][MI]Optimize the prompt information when not enabling RDR
Merge pull request !18695 from wangjunbao/master
2021-06-23 01:49:58 +00:00
Parastoo Ashtari 6ed17d52b1 Add debugger to new unified GPU runtime 2021-06-22 18:23:56 -04:00
TinaMengtingZhang 6b090679ab fix two issues: dump with input_output=1 and log type name 2021-06-22 16:13:07 -04:00
wangjunbao 2367337c2e Modify the warning message when not enabling RDR 2021-06-22 16:01:49 +08:00
wangjunbao 62e5e5bdb5 Modify the warning message when not enabling RDR 2021-06-22 14:17:14 +08:00
lanzhineng fc8ec7fc49 infer_optv3:infer optimize to find exit of branches and resolve the
endless issue.
2021-06-22 11:55:56 +08:00
ms_yan 030030deb2 delete GetEnvConfigFile func as it no longer used 2021-06-21 23:17:45 +08:00
i-robot ec999d3fa5 !18477 Modify the error message of abstract join
Merge pull request !18477 from huangbingjian/abstract_join
2021-06-18 09:10:29 +00:00
i-robot eb0bad4ad7 !16378 Export bprop func_graph to mindir
Merge pull request !16378 from YuJianfeng/bprop_mindir
2021-06-18 08:50:25 +00:00
i-robot 4dd14e7b8f !18496 overflow dump may not gen data
Merge pull request !18496 from john_tzanakakis/jt_bug_fixes
2021-06-18 07:38:15 +00:00