Tom Lane 2095206de1 Adjust btree index build to not use shared buffers, thereby avoiding the ... locking conflict against concurrent CHECKPOINT that was discussed a few weeks ago. Also, if not using WAL archiving (which is always true ATM but won't be if PITR makes it into this release), there's no need to WAL-log the index build process; it's sufficient to force-fsync the completed index before commit. This seems to gain about a factor of 2 in my tests, which is consistent with writing half as much data. I did not try it with WAL on a separate drive though --- probably the gain would be a lot less in that scenario.		2004-06-02 17:28:18 +00:00
..
attnum.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
clog.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
genam.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
gist.h	Cleanup vectors of GISTENTRY and eliminate problem with 64-bit strict-aligned	2004-03-30 15:45:33 +00:00
gistscan.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
hash.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
heapam.h	Tweak indexscan and seqscan code to arrange that steps from one page to	2004-04-21 18:24:26 +00:00
hio.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
htup.h	Replace TupleTableSlot convention for whole-row variables and function	2004-04-01 21:28:47 +00:00
ibit.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
iqual.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
itup.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
nbtree.h	Adjust btree index build to not use shared buffers, thereby avoiding the	2004-06-02 17:28:18 +00:00
printtup.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
relscan.h	Get rid of rd_nblocks field in relcache entries. Turns out this was	2004-05-08 19:09:25 +00:00
rmgr.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
rtree.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
rtscan.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
sdir.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
skey.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
slru.h	Per previous discussions, get rid of use of sync(2) in favor of	2004-05-31 03:48:10 +00:00
transam.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
tupdesc.h	Replace TupleTableSlot convention for whole-row variables and function	2004-04-01 21:28:47 +00:00
tupmacs.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
tuptoaster.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
valid.h	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
xact.h	For multi-table ANALYZE, use per-table transactions when possible	2004-05-22 23:14:38 +00:00
xlog.h	Separate out bgwriter code into a logically separate module, rather	2004-05-29 22:48:23 +00:00
xlogdefs.h	This patch is the next step towards (re)allowing fork/exec.	2003-12-20 17:31:21 +00:00
xlogutils.h	Commit the reasonably uncontroversial parts of J.R. Nield's PITR patch, to	2004-02-11 22:55:26 +00:00