Thread: pgsql: Use quicksort, not replacement selection, for external sorting.
Use quicksort, not replacement selection, for external sorting. We still use replacement selection for the first run of the sort only and only when the number of tuples is relatively small. Otherwise, the first run, and subsequent runs in all cases, are produced using quicksort. This tends to be faster except perhaps for very small amounts of working memory. Peter Geoghegan, reviewed by Tomas Vondra, Jeff Janes, Mithun Cy, Greg Stark, and me. Branch ------ master Details ------- http://git.postgresql.org/pg/commitdiff/0711803775a37e0bf39d7efdd1e34d9d7e640ea1 Modified Files -------------- doc/src/sgml/config.sgml | 39 +++ src/backend/optimizer/path/costsize.c | 6 +- src/backend/utils/init/globals.c | 1 + src/backend/utils/misc/guc.c | 10 + src/backend/utils/misc/postgresql.conf.sample | 1 + src/backend/utils/sort/tuplesort.c | 463 +++++++++++++++++++++----- src/include/miscadmin.h | 1 + 7 files changed, 430 insertions(+), 91 deletions(-)
Robert Haas <rhaas@postgresql.org> writes: > Use quicksort, not replacement selection, for external sorting. This patch evidently broke several of the Windows buildfarm critters, eg on mastodon: Build FAILED. .\src\backend\utils\sort\tuplesort.c(2373): error C2143: syntax error : missing ';' before 'type' .\src\backend\utils\sort\tuplesort.c(2377): error C2065: 'ntuples' : undeclared identifier 0 Warning(s) 2 Error(s) regards, tom lane
I wrote: > Robert Haas <rhaas@postgresql.org> writes: >> Use quicksort, not replacement selection, for external sorting. > This patch evidently broke several of the Windows buildfarm critters, On inspection, the problem is obviously "declaration in the middle of a code block". regards, tom lane
On Fri, Apr 8, 2016 at 10:47 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > I wrote: >> Robert Haas <rhaas@postgresql.org> writes: >>> Use quicksort, not replacement selection, for external sorting. > >> This patch evidently broke several of the Windows buildfarm critters, > > On inspection, the problem is obviously "declaration in the middle of > a code block". Woops. Attempted a fix. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company