database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-28 23:42:10 +03:00
Code Activity

Minor tweaking of index special-space definitions so that the various

Browse Source 
index types can be reliably distinguished by examining the special space
on an index page.  Per my earlier proposal, plus the realization that
there's no need for btree's vacuum cycle ID to cycle through every possible
16-bit value.  Restricting its range a little costs nearly nothing and
eliminates the possibility of collisions.
Memo to self: remember to make bitmap indexes play along with this scheme,
assuming that patch ever gets accepted.
This commit is contained in:
Tom Lane
2007-04-09 22:04:08 +00:00
parent 485d9ca96f
commit 56218fbc48
9 changed files with 78 additions and 64 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
src/backend/access/nbtree/nbtutils.c
View File

@ -8,7 +8,7 @@
*
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/access/nbtree/nbtutils.c,v 1.84 2007/04/06 22:33:42 tgl Exp $
* $PostgreSQL: pgsql/src/backend/access/nbtree/nbtutils.c,v 1.85 2007/04/09 22:04:01 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@ -1263,11 +1263,13 @@ _bt_start_vacuum(Relation rel)
LWLockAcquire(BtreeVacuumLock, LW_EXCLUSIVE);
/* Assign the next cycle ID, being careful to avoid zero */
do
{
result = ++(btvacinfo->cycle_ctr);
} while (result == 0);
/*
* Assign the next cycle ID, being careful to avoid zero as well as
* the reserved high values.
*/
result = ++(btvacinfo->cycle_ctr);
if (result == 0 || result > MAX_BT_CYCLE_ID)
result = btvacinfo->cycle_ctr = 1;
/* Let's just make sure there's no entry already for this index */
for (i = 0; i < btvacinfo->num_vacuums; i++)