Minor performance improvement in MultiRecordFreeSpace.

diff --git a/src/backend/storage/freespace/freespace.c b/src/backend/storage/freespace/freespace.c
index a7866745f8288590b4d46a82a1e73f97ad71c623..fec9f594a611487b1f570c07248e0725cbc29cef 100644 (file)
--- a/src/backend/storage/freespace/freespace.c
+++ b/src/backend/storage/freespace/freespace.c
@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *   $Header: /cvsroot/pgsql/src/backend/storage/freespace/freespace.c,v 1.3 2001/07/02 20:50:46 tgl Exp $
+ *   $Header: /cvsroot/pgsql/src/backend/storage/freespace/freespace.c,v 1.4 2001/07/19 21:25:37 tgl Exp $
  *
  *
  * NOTES:
@@ -419,9 +419,23 @@ MultiRecordFreeSpace(RelFileNode *rel,
         *
         * XXX we could probably be smarter about this than doing it
         * completely separately for each one.  FIXME later.
+        *
+        * One thing we can do is short-circuit the process entirely if
+        * a page (a) has too little free space to be recorded, and (b)
+        * is within the minPage..maxPage range --- then we deleted any
+        * old entry above, and we aren't going to make a new one.
+        * This is particularly useful since in most cases, all the passed
+        * pages will in fact be in the minPage..maxPage range.
         */
        for (i = 0; i < nPages; i++)
-           fsm_record_free_space(fsmrel, pages[i], spaceAvail[i]);
+       {
+           BlockNumber page = pages[i];
+           Size        avail = spaceAvail[i];
+
+           if (avail >= fsmrel->threshold ||
+               page < minPage || page > maxPage)
+               fsm_record_free_space(fsmrel, page, avail);
+       }
    }
    SpinRelease(FreeSpaceLock);
 }