projects / users / hanada / postgres.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 31f6cfb) | patch
Fix logic in lazy vacuum to decide if it's worth trying to truncate the heap.
authorHeikki Linnakangas <heikki.linnakangas@iki.fi>
Tue, 6 Jan 2009 14:56:13 +0000 (14:56 +0000)
committerHeikki Linnakangas <heikki.linnakangas@iki.fi>
Tue, 6 Jan 2009 14:56:13 +0000 (14:56 +0000)
commit4d79a504942d7e3f58133bc56087137091f3a5b7
treedbee2ea974aac96a8bc3009547ebe134c3325567tree
parent31f6cfb69077979f78fe854730c5d6e20f15a67acommit | diff
Fix logic in lazy vacuum to decide if it's worth trying to truncate the heap.
If the table was smaller than REL_TRUNCATE_FRACTION (= 16) pages, we always
tried to acquire AccessExclusiveLock on it even if there was no empty pages
at the end.

Report by Simon Riggs. Back-patch all the way to 7.4.
src/backend/commands/vacuumlazy.c diff | blob | blame | history
Hanada's development tree.