Tom Lane 2415ad9831 Teach tuplestore.c to throw away data before the "mark" point when the caller ...

is using mark/restore but not rewind or backward-scan capability.  Insert a
materialize plan node between a mergejoin and its inner child if the inner
child is a sort that is expected to spill to disk.  The materialize shields
the sort from the need to do mark/restore and thereby allows it to perform
its final merge pass on-the-fly; while the materialize itself is normally
cheap since it won't spill to disk unless the number of tuples with equal
key values exceeds work_mem.

Greg Stark, with some kibitzing from Tom Lane.

2007-05-21 17:57:35 +00:00

..

clauses.h

Get rid of some old and crufty global variables in the planner. When

2007-02-19 07:03:34 +00:00

cost.h

Teach tuplestore.c to throw away data before the "mark" point when the caller

2007-05-21 17:57:35 +00:00

geqo_copy.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_gene.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_misc.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_mutation.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_pool.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_random.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_recombination.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo_selection.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

geqo.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

joininfo.h

Refactor planner's pathkeys data structure to create a separate, explicit

2007-01-20 20:45:41 +00:00

pathnode.h

Refactor planner's pathkeys data structure to create a separate, explicit

2007-01-20 20:45:41 +00:00

paths.h

Avoid running build_index_pathkeys() in situations where there cannot

2007-04-15 20:09:28 +00:00

plancat.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

planmain.h

Teach tuplesort.c about "top N" sorting, in which only the first N tuples

2007-05-04 01:13:45 +00:00

planner.h

Expose more cursor-related functionality in SPI: specifically, allow

2007-04-16 01:14:58 +00:00

predtest.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

prep.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

restrictinfo.h

Refactor planner's pathkeys data structure to create a separate, explicit

2007-01-20 20:45:41 +00:00

subselect.h

Get rid of some old and crufty global variables in the planner. When

2007-02-19 07:03:34 +00:00

tlist.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00

var.h

Update CVS HEAD for 2007 copyright. Back branches are typically not

2007-01-05 22:20:05 +00:00