mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-01-16 12:02:42 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
77870 commits 2774 branches 343 tags 1.6 GiB
Commit graph

53 commits

Author SHA1 Message Date
Sergei Golubchik
5bf311e1e8 fix the include guards and add missing gplv2 headers 2012-02-17 12:19:38 +01:00
Sergey Petrunya
997445bc8e Make EXPLAIN better at displaying MRR/BKA: 
- "Using MRR" is no longer shown with range access.
- Instead, both range and BKA accesses will show one of the following:
  = "Rowid-ordered scan"
  = "Key-ordered scan"
  = "Key-ordered Rowid-ordered scan"
depending on whether DS-MRR implementation will do scan keys in order, rowids in order,
or both.
- The patch also introduces a way for other storage engines/MRR implementations to
  pass information to EXPLAIN output about the properties of employed MRR scans.
 2011-04-02 14:04:45 +04:00
Sergey Petrunya
e6bd643c75 MRR interface: change range_info's type from char* to range_id_t typedef. The goals are: 
- cleaner code
- ability to change from using pointers to offsets at some point
 2011-03-04 12:06:03 +03:00
Sergey Petrunya
8ef094fe4f BUG#707925: Wrong result with join_cache_level=6 optimizer_use_mrr = force (incremental, BKA join) 
- The problem was that Mrr_ordered_index_reader's interrupt_read() and resume_read() would 
  save and restore 1) index tuple  2) the rowid (as bytes returned by handler->position()).  Clustered 
  primary key columns were not saved/restored. 
  They are not explicitly present in the index tuple (i.e. table->key_info[secondary_key].key_parts 
  doesn't list them), but they are actually there, in particular 
  table->field[clustered_primary_key_member].part_of_key(secondary_key) == 1. Index condition pushdown
  code [correctly] uses the latter as inidication that pushed index condition can refer to clustered PK
  members. 

  The fix was to make interrupt_read()/resume_read() to save/restore clustered primary key members as well,
  so that we get correct values for them when evaluating pushed index condition.
[3rd attempt: remove the debugging aids, fix comments in testcase]
 2011-03-04 00:54:10 +03:00
Sergey Petrunya
9d480f6af3 BUG#670417: Diverging results in maria-5.3-mwl128-dsmrr-cpk with join buffer 
- Fixes for the second fix: take into account case where we don't need to save/restore the scan.
 2010-12-20 14:40:12 +03:00
Sergey Petrunya
09a84dc470 BUG#670417: Diverging results in maria-5.3-mwl128-dsmrr-cpk with join buffer 
Switch from "Disable identical key handling optimization when
  IndexConditionPushdown is used" approach
To 
 an approach where we save/restore index tuple and so can use index condition pushdown.
 2010-12-19 13:56:12 +03:00
Sergey Petrunya
80cbf3a1c2 Small code cleanups 2010-12-17 13:06:21 +03:00
Sergey Petrunya
c696fb6ef3 Better comments 2010-12-16 23:43:52 +03:00
Sergey Petrunya
3e633eea66 MWL#121-125 DS-MRR improvements 
- Address Monty's review feedback, portion 2
 2010-12-16 21:18:35 +03:00
Sergey Petrunya
802db7a64b - Fix compiler warning 
- Better warnings
 2010-12-15 10:45:08 +03:00
Sergey Petrunya
a86599e1e4 BUG#671340: Diverging results in with mrr_sort_keys=ON|OFF and join_cache_level=5 
- Make Mrr_ordered_index_reader() save the rowid across scan interruptions

Also
- Fix compiler warning for setup_buffer_sizes()
- Add commented key_copy/key_restore for better handling of a similar issue
  with index record being destroyed by scan interruption (which causes 
  incorrect evaluation of pushed index condition later on).
 2010-12-09 00:47:33 +03:00
Sergey Petrunya
1c7c67d584 MWL#121-125 DS-MRR improvements 
- Address Monty's review feedback, part 6: rename "handler *h" variable in all
  DS-MRR classes to something else.
 2010-12-02 16:21:52 +03:00
Sergey Petrunya
44be131cd3 MWL#121-125 DS-MRR improvements 
- Address Monty's review feedback, part 5
 2010-12-02 14:10:52 +03:00
Sergey Petrunya
40822c8149 Fix buildbot failure 2010-11-26 00:30:39 +03:00
Sergey Petrunya
92442fc598 Merge MWL#121-125 DS-MRR improvements 
- Address Monty's review feedback, part 4
 2010-11-25 14:35:21 +03:00
Sergey Petrunya
a6c1d56e08 MWL#121-125 DS-MRR improvements 
- Address Monty's review feedback, part 1
- Fix buildbot failure
 2010-11-22 19:34:03 +03:00
Sergey Petrunya
2ec43747f5 Merge MWL#121-125 DS-MRR improvements 
- Address Monty's review feedback, part 1
 2010-11-22 17:08:22 +03:00
Sergey Petrunya
0fb342ba56 BUG#670417: Diverging results in maria-5.3-mwl128-dsmrr-cpk with join buffer, attempt 4 
- Disable identical key handling optimization when
  IndexConditionPushdown is used
 2010-11-08 20:37:01 +03:00
Sergey Petrunya
197c99427d DS-MRR improvements 
- Code cleanup
- Always propagate the error code we got from storage engine all the way up
 2010-11-08 15:15:50 +03:00
Sergey Petrunya
c46450de72 Code cleanup 2010-11-03 00:09:28 +03:00
Sergey Petrunya
1654780f2d Code cleanup 2010-11-02 23:25:35 +03:00
Sergey Petrunya
b76a8595c6 Re-initialization reworked 2010-11-01 13:52:10 +03:00
Sergey Petrunya
d9a8dd22b2 DS-MRR/CPK improvements: correct buffer exhaustion handling 2010-10-31 22:00:15 +03:00
Sergey Petrunya
1e82320d27 DS-MRR improvements 
- buildbot test failure fixes: don't try to get more records from SimpleIndexReader
  if we've already got EOF for it.
 2010-10-29 13:42:48 +04:00
Sergey Petrunya
a88bd5cf56 DS-MRR improvements: 
- address the review feedback (class/var renames)
- fix wrong test result (bug in buffer refill logic)
 2010-10-28 20:48:16 +04:00
Sergey Petrunya
d8efc3b155 DS-MRR improvements: address review feedback for R3 version of the patch 2010-10-26 15:35:13 +04:00
Sergey Petrunya
ac8a79b944 Code cleanup 2010-10-04 10:31:40 +04:00
Sergey Petrunya
5156457506 Address review feedback 
- One iterator class
- Switch back from state automaton into two-nested-iterators approach..
 2010-10-03 14:48:42 +04:00
Sergey Petrunya
22d5323fac DS-MRR/CPK improvements: more of addressing review feedback 2010-10-01 15:54:35 +04:00
Sergey Petrunya
61f26f0c62 DS-MRR improvements: address review feedback 
- change dsmrr_next_from_index() to a switch-based state automaton-like structure.
 2010-09-28 20:20:09 +04:00
Sergey Petrunya
4f56acb676 Better comments, move Lifo_buffer to separate file. 2010-09-28 12:19:50 +04:00
Sergey Petrunya
3066c37718 DS-MRR improvements: review feedback 
- Switch from one bi-directional buffer class to two 
  virtual inheritance-based forward and backward buffer classes.
 2010-09-21 20:19:54 +04:00
Sergey Petrunya
51f9097608 More comments 2010-09-20 23:13:28 +04:00
Sergey Petrunya
2121ab1eb4 DS-MRR improvements: remove write_size/read_size, have the same size 
for writing and reading
 2010-09-20 13:23:51 +04:00
Sergey Petrunya
189555f39a DS-MRR improvements: more code cleanup 
- better comments
- rename variables to better reflect their meaning
 2010-09-20 13:02:17 +04:00
Sergey Petrunya
18a348503a DS-MRR improvements: better comments, use symbolic name instead of +1/-1 constants. 2010-09-19 01:05:47 +04:00
Sergey Petrunya
499b142ad5 BUG#628785: multi_range_read.cc:430: int DsMrr_impl::dsmrr_init(): Assertion `do_sort_keys || do_rowid_fetch' failed 
- Make Ds_MrrImpl::check_cpk_scan() follow the execution code' logic: don't
  do MRR scans on clustered PK when mrr_sort_keys=off.
 2010-09-15 16:58:01 +04:00
Sergey Petrunya
188de43db1 - Fixed BUG#623285: Crash in quick_range_seq_next() in maria-5.3-dsmrr-cpk 
- Relax overly-strict assert added in previous commit
- Run generic MRR testsuite for Maria engine, too.
 2010-09-13 19:45:41 +04:00
Sergey Petrunya
ac3756a1d0 Debug measure: trash unused buffer space 2010-09-12 18:28:19 +04:00
Sergey Petrunya
9b04caffd4 Commit for buildbot checks 2010-09-10 20:48:11 +04:00
Sergey Petrunya
dbc63bed22 MWL#121-125: DS-MRR improvements 
- Address review feedback, step 1
 2010-09-05 14:32:14 +04:00
Sergey Petrunya
937db4bff4 - Remove out-of-date comments 
- Make testcase stable
 2010-08-14 22:35:50 +04:00
Sergey Petrunya
c964cb1b62 key/rowid buffer overflow fixes for various tricky cases. 2010-08-14 18:56:37 +04:00
Sergey Petrunya
d098596ba5 Fix a number of problems with reverse buffer use 2010-08-12 23:59:29 +04:00
Sergey Petrunya
8d07c16ad9 Do dynamic buffer growing/shrinking. 2010-08-12 21:18:41 +04:00
Sergey Petrunya
8dc81f47ef Use reverse/backwards buffer for keys (now works) 
- don't allocate space for rowid buffer when we don't really need it.
- fix buffer iterator
 2010-08-11 14:54:34 +04:00
Sergey Petrunya
e25db795bd DS-MRR, key-ordered retrievals: commit for buildbot 2010-08-08 11:13:54 +04:00
Sergey Petrunya
7391175d7e DS-MRR support improvements (MWL#123, MWL#124, MWL#125) 
- Let DS-MRR/CPK take advantage of materialized keys, when possible
 2010-07-18 01:05:44 +04:00
Sergey Petrunya
e0999cdf7c DS-MRR support improvements (MWL#123, MWL#124, MWL#125) 
- Lots of TODO comments
- add mrr_sort_keys flag to @@optimizer_switch
- [from Igor] SQL layer part passes HA_MRR_MATERIALIZED_KEYS flag
- Don't call rnd_pos() many times in a row if sorted rowid buffer
  has the same rowid value for multiple consequive (rowid, range_id) pairs.
 2010-07-17 18:03:50 +04:00
Sergey Petrunya
488d352a66 MWL#121: DS-MRR support for clustered primary keys 
- Remove back key_parts from multi_range_read_init() parameters
- Related code simplification/cleanup
 2010-06-22 23:26:11 +04:00