Howard Chu
b0851a13af
ITS#8316, 8321
2015-11-23 01:51:45 +00:00
Howard Chu
d78ffc9517
ITS#8321 reorganize page_split fixups
...
DUPFIXED fixups needed to occur after separator update.
MDB_RESERVE handling moved after split fixup.
2015-11-23 01:50:29 +00:00
Howard Chu
8fdf79600a
ITS#8321 simplify page_split fixups
2015-11-23 01:50:17 +00:00
Howard Chu
7a76ded030
ITS#8321 track temporary cursors
...
In rebalance/split operations, temporary cursors need to be visible
to propagate fixups
2015-11-23 01:50:03 +00:00
Howard Chu
5c7b84b465
ITS#8321 fix mdb_cursor_put
...
Ignore sub-cursors that shouldn't be fixed up
2015-11-23 01:49:49 +00:00
Howard Chu
94831f7c3f
ITS#8321 fix mdb_cursor_shadow()
...
Set a valid txn so that cursor fixup code works on the shadows
2015-11-23 01:49:37 +00:00
Howard Chu
0ec3967e1d
ITS#8321 fix mdb_cursor_chk()
...
It was reporting spurious errors due to uninit'd cursors
2015-11-23 01:49:24 +00:00
Howard Chu
8773a08c43
ITS#8321 don't skip fixups on splitting cursors
...
Adjustments can't be skipped, in recursive calls each level must
fixup their own level.
2015-11-23 01:49:10 +00:00
Howard Chu
91dc62506b
ITS#8321 Fix del/dupsort
...
When deleting a dupsort key, if other cursors pointed at that key,
set them to uninit'd, not EOF. They no longer have anything to
point at.
2015-11-23 01:48:59 +00:00
Howard Chu
94e8009ca4
ITS#8300 more for prev commit
...
Just tell explicitly which direction we moved/merged from
2015-11-23 01:47:39 +00:00
Howard Chu
ba85adb52d
Silence some valgrind uninit warnings
2015-11-23 01:47:39 +00:00
Howard Chu
bfe2088946
Refix root split check from 5da67968af
2015-11-23 01:47:39 +00:00
Oskari Timperi
f13b971960
ITS#7992 assume Windows paths are UTF-8
2015-11-23 01:47:39 +00:00
Howard Chu
28b57ba8b9
ITS#8316 cursor fixup in page_merge
...
Deleting the merged page requires bumping down other ki's
of the page's parent.
2015-11-23 01:47:39 +00:00
Howard Chu
657dbcc811
ITS#8312, 8313, 8315
2015-11-19 17:29:42 +00:00
Howard Chu
1edb0e3a42
ITS#8315 fix ovpage_free
...
Keep dirty_room sync'd with dirty_list
2015-11-19 17:28:33 +00:00
Howard Chu
2fb8219aa3
ITS#8313 more for ITS#8062
...
dummy flags must be init'd due to 3d46d550
2015-11-19 17:28:23 +00:00
Hallvard Furuseth
db510359c9
ITS#8312 Fix loose pages in commit(nested txn)
2015-11-19 17:28:15 +00:00
Howard Chu
27f71b4dd3
ITS#8311
2015-11-18 10:27:50 +00:00
Howard Chu
71c07f0d60
ITS#8311 add comment
2015-11-18 10:24:41 +00:00
Howard Chu
f9d3dc8293
ITS#8311 fix page_split from update_key
...
Check for top of stack. Usually the cursor only has height 1 when
calling page_split, but not always.
2015-11-18 10:05:28 +00:00
Hallvard Furuseth
a9877f783a
ITS#7209
2015-11-17 18:49:31 +01:00
Hallvard Furuseth
2a839d5dfa
ITS#7209 Default MDB_DSYNC = O_SYNC if no O_DSYNC
2015-11-17 18:37:33 +01:00
Howard Chu
f6b7277e9a
ITS#8310
2015-11-13 09:28:36 +00:00
Howard Chu
a90415f301
ITS#8310 fix mdb_rebalance cursor fixup
...
When collapsing root, must adjust entire cursor stack
2015-11-13 09:25:55 +00:00
Howard Chu
79b12bbd7b
Tweak robust mutex detection for glibc
2015-11-11 11:51:07 +00:00
Howard Chu
4e1ec79f0e
ITS#8304
2015-11-09 20:11:10 +00:00
Howard Chu
5d8f9ddf7e
ITS#8304 fix page_merge
...
Was using the pre-touch dst page pointer instead of the touched page
2015-11-09 20:10:51 +00:00
Howard Chu
712942d392
ITS#8300 more for node_move
...
When moving a node from one page to another, make sure other cursors'
parent index gets adjusted too.
2015-11-06 10:19:49 +00:00
Howard Chu
7d022e05a8
ITS#8300 simplify
2015-11-04 22:46:04 +00:00
Howard Chu
0db71b9f1a
ITS#8300 more for node_move fixups
...
When moving a node from the right neighbor, a different
adjustment is needed.
2015-11-04 22:31:01 +00:00
Howard Chu
e339fced30
ITS#8300
2015-11-04 21:03:22 +00:00
Howard Chu
9fc7c94aeb
ITS#8300 fix node_move
...
Don't adjust other cursors when we added a node on the right.
2015-11-04 21:01:30 +00:00
Howard Chu
af45cf3a04
ITS#8300 fix rebalance after node_move
...
ITS#8258, ITS#7829 fixes checked parent index to see if we were moving
from a left neighbor. Should have just checked to see if current index
was 0, meaning we added on the left. (Parent index may not tell us
anything meaningful after a nested rebalance.)
2015-11-04 20:38:30 +00:00
Howard Chu
4359ffe08f
ITS#8238, #8299
2015-11-04 18:17:50 +00:00
Howard Chu
17aab561f2
ITS#8238 fix DUPFIXED page_split
...
Parent mc_ki wasn't adjusted if new_indx was > split point
2015-11-04 18:11:12 +00:00
Howard Chu
c62b77b661
ITS#8299 fix page_merge cursor fixup
...
The parent's mc_ki has changed. We need to fix that up as well
in other cursors.
2015-11-04 17:05:55 +00:00
Howard Chu
caca4d8be6
Use MDB_USE_ROBUST not MDB_NO_ROBUST
2015-10-31 02:27:31 +00:00
Howard Chu
447f552756
Add MDB_NO_ROBUST to disable Robust Mutexes
2015-10-31 02:12:31 +00:00
Howard Chu
550df2a583
More misc updates
2015-10-26 20:18:19 +00:00
Hallvard Furuseth
f1e07f9d14
Fix typo in mdb_tassert().
...
No effect on current code.
2015-10-25 14:40:27 +01:00
Pavel Medvedev
56adc467a6
ITS#8069 char* strings on Windows are ASCII
2015-10-25 10:56:39 +01:00
Pavel Medvedev
22958b0f9e
ITS#8067 add ssize_t typedef for MSVC
2015-10-25 10:56:39 +01:00
Hallvard Furuseth
de5b689308
Simpler mdb_env_write_meta()
2015-10-25 10:56:39 +01:00
Hallvard Furuseth
bde5c231bc
Kill magic numbers for NUM_METAS, databases
2015-10-25 10:56:38 +01:00
Hallvard Furuseth
8901d9cfbc
ITS#7377 Catch mdb_put() to blocked/read-only txns
...
...early enough that txn state is left unchanged.
2015-10-25 10:55:22 +01:00
Hallvard Furuseth
01342bd5d2
ITS#7377 Always notice env error on txn startup.
...
Move the check to the end of txn startup. Catches env
breakage which happens while the new txn waits for a lock.
2015-10-25 10:55:22 +01:00
Hallvard Furuseth
602c978761
Cleanup: Drop !mt_dbxs hack, use MDB_TXN_FINISHED
2015-10-25 10:55:22 +01:00
Hallvard Furuseth
a3ae2e7db1
Catch most uses of finished/parent txns.
...
* Add MDB_TXN_FINISHED, MDB_TXN_HAS_CHILD, MDB_TXN_BLOCKED.
* Clear mt_numdbs in writers, for TXN_DBI_EXIST() to catch.
We already do in readers.
2015-10-25 10:55:22 +01:00
Hallvard Furuseth
9266843fa6
Move code into mdb_txn_end(). Was mdb_txn_reset0.
...
Side effects:
* Clean txn up a bit even before freeing it.
* Tweak DEBUG output at txn end. Add DEBUG after commit(writer).
2015-10-25 10:55:22 +01:00
