]> granicus.if.org Git - postgresql/commit
Adjust tuplesort.c based on the fact that we never use the OS's qsort().
authorRobert Haas <rhaas@postgresql.org>
Thu, 26 Jan 2012 19:43:28 +0000 (14:43 -0500)
committerRobert Haas <rhaas@postgresql.org>
Thu, 26 Jan 2012 19:43:28 +0000 (14:43 -0500)
commitc5a03256c725c09c32a5c498bd7c8799ed3ec2a0
tree96a68bbac3a09251c0b5d374117e7c493aa5c092
parent2d1371d3ee5cf7e96d16fb503c27e060df9497f7
Adjust tuplesort.c based on the fact that we never use the OS's qsort().

Our own qsort_arg() implementation doesn't have the defect previously
observed to affect only QNX 4, so it seems sufficiently to assert that
it isn't broken rather than retesting.  Also, update a few comments to
clarify why it's valuable to retain a tie-break rule based on CTID
during index builds.

Peter Geoghegan, with slight tweaks by me.
src/backend/utils/sort/tuplesort.c