Commit Graph

21 Commits

Author SHA1 Message Date
9216735cfa [New Featrue] Support Vectorization Execution Engine Interface For Doris (#6329)
1. FE vectorized plan code
2. Function register vec function
3. Diff function nullable type
4. New thirdparty code and new thrift struct
2021-08-11 14:54:06 +08:00
98e80aa65e [refactor] Replace boost::function with std::function (#5700)
Replace boost::function with std::function
2021-05-09 22:00:48 +08:00
6fedf5881b [CodeFormat] Clang-format cpp sources (#4965)
Clang-format all c++ source files.
2020-11-28 18:36:49 +08:00
10e1e29711 Remove header file common/names.h (#4945) 2020-11-26 17:00:48 +08:00
75e0ba32a1 Fixes some be typo (#4714) 2020-10-13 09:37:15 +08:00
10f822eb43 [MemTracker] make all MemTrackers shared (#4135)
We make all MemTrackers shared, in order to show MemTracker real-time consumptions on the web.
As follows:
1. nearly all MemTracker raw ptr -> shared_ptr
2. Use CreateTracker() to create new MemTracker(in order to add itself to its parent)
3. RowBatch & MemPool still use raw ptrs of MemTracker, it's easy to ensure RowBatch & MemPool destructor exec 
     before MemTracker's destructor. So we don't change these code.
4. MemTracker can use RuntimeProfile's counter to calc consumption. So RuntimeProfile's counter need to be shared 
    too. We add a shared counter pool to store the shared counter, don't change other counters of RuntimeProfile.
Note that, this PR doesn't change the MemTracker tree structure. So there still have some orphan trackers, e.g. RowBlockV2's MemTracker. If you find some shared MemTrackers are little memory consumption & too time-consuming, you could make them be the orphan, then it's fine to use the raw ptr.
2020-07-31 21:57:21 +08:00
d0fe7e4d94 [Profile] Make running profile clearer and more intuitive to improve usability (#3405)
This CL mainly made the following modifications:
    1. Delete Invalid MemoryUsed Counter and Add PeakMemUsage in each exec node and datastreamsender
    2. Add intent in child execnode profile,make it is easily to know the relationship between execnode
    3. Del _is_result_order we not support any more in olap_scan_node.h and olap_scan_node.cpp
    4. Add scan_disk method to olap_scanner to fix the counter _num_disks_accessed_counter
    5. Now we do not use buffer pool to read and write disk, so annotation eadio counter and 
    6. Delete the MemUsed counter in exec node.
2020-04-30 14:57:21 +08:00
a340bc7a00 Remove unused LLVM related codes of directory:be/src/runtime (#2910) (#2985)
Remove unused LLVM related codes of directory (step 4):be/src/runtime (#2910)

there are many LLVM related codes in code base, but these codes are not really used.
The higher version of GCC is not compatible with the LLVM 3.4.2 version currently used by Doris.
The PR delete all LLVM related code of directory: be/src/runtime
2020-02-25 13:47:20 +08:00
839ec45197 Remove llvm relative code from be/src/exec (#2955)
Remove unused LLVM related codes of directory:be/src/exec (#2910)

there are many LLVM related codes in code base, but these codes are not really used.
The higher version of GCC is not compatible with the LLVM 3.4.2 version currently used by Doris.
The PR delete all LLVM related code of directory: be/src/exec.
2020-02-20 20:43:26 +08:00
852046de29 Fix incompatibility with arm architecture in olap #2645 (#2682) 2020-01-07 19:16:10 +08:00
42395d2455 Change Null-safe equal operator from cross join to hash join (#2156)
* Change Null-safe equal operator from cross join to hash join
ISSUE-2136

This commit change the join method from cross join to hash join when the equal operator is Null-safe '<=>'.
It will improve the speed of query which has the Null-safe equal operator.
The finds_nulls field is used to save if there is Null-safe operator.
The finds_nulls[i] is true means that the i-th equal operator is Null-safe.
The equal function in hash table will return true, if both val and loc are NULL when finds_nulls[i] is true.
2019-11-08 12:43:48 +08:00
9f5e5717d4 Unify the msg of 'Memory exceed limit' (#1737)
The new msg of limit exceed: "Memory exceed limit. %msg, Backend:%ip, fragment:%id Used:% , Limit:%. xxx".
This commit unifies the msg of 'Memory exceed limit' such as check_query_state, RETURN_IF_LIMIT_EXCEEDED and LIMIT_EXCEEDED.
2019-09-03 10:42:16 +08:00
9d03ba236b Uniform Status (#1317) 2019-06-14 23:38:31 +08:00
a2b299e3b9 Reduce UT binary size (#314)
* Reduce UT binary size

Almost every module depend on ExecEnv, and ExecEnv contains all
singleton, which make UT binary contains all object files.

This patch seperate ExecEnv's initial and destory to anthor file to
avoid other file's dependence. And status.cc include debug_util.h which
depend tuple.h tuple_row.h, and I move get_stack_trace() to
stack_util.cpp to reduce status.cc's dependence.

I add USE_RTTI=1 to build rocksdb to avoid linking librocksdb.a

Issue: #292

* Update
2018-11-15 16:17:23 +08:00
37b4cafe87 Change variable and namespace name in BE (#268)
Change 'palo' to 'doris'
2018-11-02 10:22:32 +08:00
2868793b6b Change license to Apache License 2.0 (#262) 2018-11-01 09:06:01 +08:00
2419384e8a push 3.3.19 to github (#193)
* push 3.3.19 to github

* merge to 20ed420122a8283200aa37b0a6179b6a571d2837
2018-05-15 20:38:22 +08:00
5de798fdd6 Merge code to github (#187)
* merge to 95787f8be1fd0ff215708fb0f49997b632876586
* Bugs fixed
2018-03-23 14:04:55 +08:00
a7fe56322d aggregation with limit bug (#152) 2017-12-07 15:02:10 +08:00
ddf6c38b0a aggregation with limit bug (#151) 2017-12-07 14:58:18 +08:00
e2311f656e baidu palo 2017-08-11 17:51:21 +08:00