5f68293475
explain mdl_midl_sort() istack size
13 years ago
2d1f3b7ed3
Fix memn(r)cmp of key sizes > INT_MAX. Simplify.
13 years ago
69e53cf700
Check some mdb integer type assumptions
13 years ago
88da18cccf
Put MDB_node.<mn_lo,mn_hi> in host byte order.
13 years ago
c5dad7a6d0
C90 compatibility cleanup in mdb.
...
Fix void* pointer arithmetic in cintcmp().
Fix invalid ';'s in declarations after possibly-empty macros.
13 years ago
3dc8fa491a
Cleanup mdb macros.
...
Parenthesize. Rename GetPageSize -> GET_PAGESIZE since it does not
behave like a function (it has a non-pointer output argument).
13 years ago
ec2d82f79a
libmdb/.gitignore += test[45], Doxygen output
13 years ago
acdc248f57
If put(NOOVERWRITE) exists, return existing data
13 years ago
a7edb95e64
Minor typos, additional clarification
13 years ago
0c487538cf
Partial revert, fix MDB_page definition
13 years ago
9ac8638d9d
More docs
13 years ago
bed3123b82
More docs
13 years ago
1b69295a48
Speedup cintcmp
...
Now that we know we're always 2-byte aligned...
13 years ago
dac3fae3b5
Fix previous commit
13 years ago
8678fef597
Forgot to commit this
13 years ago
47c1218916
Forgot to commit this
13 years ago
2e3bc39fa9
Further cursor-based restructuring
...
Get rid of all of the redundant MDB_*page structs and other cruft.
13 years ago
976a150973
Cleanup unused code
13 years ago
1f8c3369b9
Fix mdb_move_node again
13 years ago
f9c8796d0b
Fix search_node, add cintcmp
13 years ago
5e6a864513
Guarantee nodes are 2-byte aligned
...
Also fix xcursor_init1 md_dirty flag
Also another mdb_cmp refactoring fix
13 years ago
293df78b2b
Change MDB_node to depend on 2-byte alignment
13 years ago
807841b1aa
Fix for mdb_cmp refactoring
13 years ago
963c421a97
Minor cleanup
13 years ago
2cde97ae94
Use mdb_midl_append/sort instead of _insert
13 years ago
70a4f6f29d
mdb_cmp refactoring
13 years ago
7bb14a2e79
Cursor stuff working, no major perf gain
13 years ago
1843fe0b53
Cursors in-progress, broken
13 years ago
8d89a53165
Use cursor->mc_flags
13 years ago
7f46832ec5
Begin reimplementing basic ops as cursor ops
13 years ago
f6edbd7793
Add MDB_{FIRST,LAST}_DUP for cursor_get
13 years ago
9d821c26c3
Fix mdb_split, fix MDB_GET_BOTH
13 years ago
e91dae3ed5
Cleanup midl, start documenting internals
13 years ago
b30d0266e7
Tweak docs
13 years ago
30736a0ff5
Ugly-as-sin initial Windows port
13 years ago
2510bea718
API cleanup, doxygen docs for public API
13 years ago
0f459dd773
Cleanup, fix freelist alloc
...
Don't allow new pages for free list to come from the free list.
Otherwise a nasty data self-reference occurs that is too much
trouble to unwind.
13 years ago
f7aa0d5e4a
Tweak locks, fix race conditions
13 years ago
d27a7754f3
Reduce excessive memcpy in mdb_split
13 years ago
027e1daa10
Fix mdb_split, tweak split_indx if space is tight
13 years ago
01b9fc59fb
Fix update of big data items
13 years ago
13c663f666
Fix free page reclaim and dbs table update
...
Can re-use free pages 1 txn earlier
Must update all dbs tables on txn commit
13 years ago
94718fab00
More debug, fix MDB_DEL_DUP
13 years ago
ee69051a7c
More debug output
13 years ago
d54eb7dd81
Add txn_reset / txn_renew
13 years ago
590c728044
First cut at DUPFIXED support
...
Also in cursor_set, check the current page (if already set from
before) before starting over again from the root.
13 years ago
720c4b0a3d
Set MINKEYS to 2
...
For better space utilization
13 years ago
66a647e55d
Keep re-using dpages instead of freeing them
13 years ago
ca496bd357
Avoid cast-away-const for errstr[]
13 years ago
5b3b522313
Wrap O_DSYNC in MDB_DSYNC.
...
If O_DSYNC is not defined, that may be due to poor compiler options.
We could fall back to the less efficient O_SYNC, but it seems better
to let the user learn of the problem and give better compiler options.
13 years ago
Hallvard Furuseth