mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-02-02 03:51:50 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
5794 commits 2793 branches 343 tags 1.7 GiB
Commit graph

5794 commits

This branch
This branch
All branches
Author SHA1 Message Date
John Esmet
ed9b5fc676 fixes #140 Pre-size the fifo before appending data during nonleaf 
serialization. Assume the fifo is about 1kb larger than the read buffer
(in practice, the fifo is 5 bytes smaller than the read buffer).
 2014-01-21 13:02:59 -05:00
John Esmet
84fa905d8d fixes #130 Create basement nodes and non-leaf message trees using 
create_no_array() so that deserialization does not malloc/free needlessly.
 2014-01-21 12:59:58 -05:00
John Esmet
d000732ab4 fixes #151 Do not send commit messages for fractal tree update commands 2014-01-21 12:59:46 -05:00
John Esmet
2e190e0631 fixes #152 Add accounting for long input buffer wait stalls in the 
logger
 2014-01-21 12:59:34 -05:00
John Esmet
f7c1d3f9ff fixes #154 Remove toku_fair_rwlock 2014-01-20 16:53:17 -05:00
Rich Prohaska
0492446818 #147 dont hold locktree manager mutex when escalating 2014-01-14 19:14:18 -05:00
Rich Prohaska
c8993dec4d #141 identify a big transaction and have it run lock escalation 2014-01-13 11:29:52 -05:00
John Esmet
96a2a6fc96 fixes #134 Add the get/set/change_fanout API 2014-01-09 14:34:23 -05:00
Rich Prohaska
e3b21c8f47 #139 disable the lock escalation thread. new algorithm coming in #141 2014-01-08 07:34:54 -05:00
John Esmet
23fa28bbbe fixes #135 Don't hold the write list lock while evicting a node 2014-01-07 17:29:07 -05:00
John Esmet
96afd28faa fixes #131 Don't hold the pair's mutex while running partial eviction. 2014-01-07 15:10:51 -05:00
Rich Prohaska
dddda9d622 #137 fix osx build problem 2014-01-07 03:57:20 -05:00
Rich Prohaska
f64edcc463 #137 error out env->open when worker threads can not be created 2014-01-06 16:58:22 -05:00
John Esmet
bd5111d7c5 refs #123 Touch the pair regardless of TOKUDB_TRY_AGAIN, to make the 
clock behavior more consistent with the rest of the code.
 2013-12-31 13:23:52 -05:00
John Esmet
8664a78ebb refs #123 Touch the clock on a successful pin in maybe_pin_pair() 2013-12-31 12:04:39 -05:00
Leif Walsh
44002ac2b1 Merge commit 'bafb665' 2013-12-30 11:08:17 -05:00
Leif Walsh
21830bbd68 fixed compilation with clang due to uninteresting -Wignored-attributes Tokutek/mongo#869 2013-12-30 11:06:55 -05:00
Leif Walsh
d1bef7992e fixed osx compilation breakage due to format strings 2013-12-23 19:23:20 -05:00
Leif Walsh
9db054d403 removed strict valgrind requirement #122 Tokutek/mongo#869 
fixes #122
 2013-12-23 19:22:35 -05:00
Leif Walsh
c99ee6b919 fixed some invalid parameter passing discovered by clang Tokutek/mongo#869 
TXNID_NONE is not the same as (TOKUTXN) 0
 2013-12-21 19:01:09 -05:00
Leif Walsh
5faf2143a8 silenced tokudb.data warning if being built as a component Tokutek/mongo#869 2013-12-21 19:00:29 -05:00
Leif Walsh
9f59d06843 added cpack component identifications to all install targets Tokutek/mongo#869 2013-12-20 17:58:48 -05:00
Leif Walsh
47b4a0fb4c Merge branch 'osx' 2013-12-17 18:29:10 -05:00
Leif Walsh
b234a26410 fixed some minor issues OSX complains about Tokutek/mongo#859 2013-12-17 18:29:06 -05:00
Rich Prohaska
ad2f063d99 #119 translate CRNL to NL 2013-12-16 15:12:51 -05:00
Zardosht Kasheff
6721a2c1e7 Merge branch 'bugs/115' 2013-12-09 15:46:56 -05:00
Zardosht Kasheff
41364a4ba8 refs #115, don't allow mempools to grow unbounded by restricting their size in 
omt_compress_kvspace.
 2013-12-09 15:46:49 -05:00
Zardosht Kasheff
046f70889d Merge branch 'bugs/113' 2013-12-06 15:01:39 -05:00
Zardosht Kasheff
439e846e3d refs #113, add a callback function to the cursor to check for interrupts. 
This allows callers to prevent a cursor that churns through many many committed
deletes to run amok.
 2013-12-06 15:01:28 -05:00
Leif Walsh
a0aa7c8558 changed logformat to link statically with portability #112 
fixes #112
 2013-12-04 11:26:20 -05:00
Rich Prohaska
dbfddf4e0d Tokutek/mariadb#12 rebase to mariadb-5.5.34 2013-11-28 09:52:08 -05:00
Rich Prohaska
78bfa5e78c #107 use frwlock as checkpoint_safe_lock 2013-11-18 09:31:36 -05:00
Rich Prohaska
be83087a3c #107 check if various rwlocks are fair WRT multiple writers 2013-11-18 08:19:39 -05:00
Rich Prohaska
8ab796c431 #90 debug long fsync system calls 2013-11-11 08:02:31 -05:00
Leif Walsh
f5ca05e283 Merge pull request #100 from Tokutek/mongo/650 
remove fullhash from rollback_log_nodes

fixes Tokutek/mongo#650
 2013-11-05 13:28:58 -08:00
Yoni Fogel
a3dec7add8 Refs Tokutek/mongo#650 simplify test, fix comments 2013-11-05 08:33:08 -08:00
Leif Walsh
45845261ba added recover-child-rollback to run.stress-tests.py Tokutek/mongo#650 2013-11-04 22:17:02 -05:00
Yoni Fogel
9377587bdf refs Tokutek/mongo#650 Add regression test. 2013-11-04 22:17:02 -05:00
Yoni Fogel
4d89b184e5 Add case for new version 2013-11-04 22:17:02 -05:00
Yoni Fogel
6120239edf Refs Tokutek/mongo#650 Remove fullhash from rollback_log_nodes; fixes issue with large (or childful) transactions during recovery due to hash_id and filenum being decoupled 2013-11-04 22:17:01 -05:00
zkasheff
b5d8ed6066 refs Tokutek/mongo#645, change signature of hot_optimize to have output variable 
of number of loops run
 2013-10-21 16:02:14 -04:00
Rich Prohaska
293a684e92 #87 add callback to allow loader memory size to be a session variable 2013-10-14 14:32:35 -04:00
Rich Prohaska
1ed8c69b01 #81 export an API to set a get wait time callback 2013-10-10 14:07:51 -04:00
Rich Prohaska
d817d01ab7 #81 move the lock wait time to the wait method 2013-10-10 10:00:35 -04:00
John Esmet
7944bc8a34 Fix build 2013-10-09 11:57:53 -04:00
Rich Prohaska
5dc1ef2f4f #56 simply launch of lock escalator and write a slightly better test 2013-10-08 14:52:34 -04:00
Zardosht Kasheff
27f5720d4a refs #82, fix CMakeLists.txt 2013-10-07 10:40:40 -04:00
Zardosht Kasheff
4f74e29926 refs #82, remove checkpoint_1.cc for TokuDB 7.1.0, may bring it back fixed later. That is still open 2013-10-07 10:35:05 -04:00
Zardosht Kasheff
548bc646b2 refs #78 for minicrons with a period of longer than one second (checkpoints), 
change minicron to count the period from the beginning of
the callback's execution and not the end. For checkpoints, this makes them
reliably start every 60 seconds.
 2013-10-07 10:17:03 -04:00
Zardosht Kasheff
6e60da0886 refs #84, remove some shared variables from logger 2013-10-07 10:16:03 -04:00