pg.ddx.io pgsql-committers@postgresql.org mailing list archivehelp / color / mirror / Atom feed
pgsql: Grab the low-hanging fruit from forcing USE_FLOAT8_BYVAL to true 2+ messages / 1 participants [nested] [flat]
* pgsql: Grab the low-hanging fruit from forcing sizeof(Datum) to 8. @ 2025-08-13 21:18 Tom Lane <tgl@sss.pgh.pa.us> 0 siblings, 0 replies; 2+ messages in thread From: Tom Lane @ 2025-08-13 21:18 UTC (permalink / raw) To: pgsql-committers@lists.postgresql.org Grab the low-hanging fruit from forcing sizeof(Datum) to 8. Remove conditionally-compiled code for smaller Datum widths, and simplify comments that describe cases no longer of interest. I also fixed up a few more places that were not using DatumGetIntXX where they should, and made some cosmetic adjustments such as using sizeof(int64) not sizeof(Datum) in places where that fit better with the surrounding code. One thing I remembered while preparing this part is that SP-GiST stores pass-by-value prefix keys as Datums, so that the on-disk representation depends on sizeof(Datum). That's even more unfortunate than the existing commentary makes it out to be, because now there is a hazard that the change of sizeof(Datum) will break SP-GiST indexes on 32-bit machines. It appears that there are no existing SP-GiST opclasses that are actually affected; and if there are some that I didn't find, the number of installations that are using them on 32-bit machines is doubtless tiny. So I'm proceeding on the assumption that we can get away with this, but it's something to worry about. (gininsert.c looks like it has a similar problem, but it's okay because the "tuples" it's constructing are just transient data within the tuplesort step. That's pretty poorly documented though, so I added some comments.) Author: Tom Lane <tgl@sss.pgh.pa.us> Reviewed-by: Peter Eisentraut <peter@eisentraut.org> Discussion: https://postgr.es/m/1749799.1752797397@sss.pgh.pa.us Branch ------ master Details ------- https://git.postgresql.org/pg/commitdiff/6aebedc38497ecebda22caf61bcadf78a8331b52 Modified Files -------------- doc/src/sgml/xfunc.sgml | 3 +- src/backend/access/gin/gininsert.c | 5 +- src/backend/access/gist/gistproc.c | 10 +-- src/backend/access/nbtree/nbtcompare.c | 20 ------ src/backend/catalog/pg_type.c | 4 +- src/backend/utils/adt/mac.c | 27 +++----- src/backend/utils/adt/network.c | 54 ++++----------- src/backend/utils/adt/numeric.c | 117 ++++----------------------------- src/backend/utils/adt/timestamp.c | 21 ------ src/backend/utils/adt/uuid.c | 6 +- src/backend/utils/adt/varlena.c | 27 +++----- src/backend/utils/sort/tuplesort.c | 8 --- src/include/access/gin_tuple.h | 4 +- src/include/access/spgist_private.h | 14 ++-- src/include/access/tupmacs.h | 15 ++--- src/include/port/pg_bswap.h | 17 +---- src/include/utils/sortsupport.h | 4 -- 17 files changed, 72 insertions(+), 284 deletions(-) ^ permalink raw reply [nested|flat] 2+ messages in thread
* pgsql: Grab the low-hanging fruit from forcing USE_FLOAT8_BYVAL to true @ 2025-08-13 21:18 Tom Lane <tgl@sss.pgh.pa.us> 0 siblings, 0 replies; 2+ messages in thread From: Tom Lane @ 2025-08-13 21:18 UTC (permalink / raw) To: pgsql-committers@lists.postgresql.org Grab the low-hanging fruit from forcing USE_FLOAT8_BYVAL to true. Remove conditionally-compiled code for the other case. Replace uses of FLOAT8PASSBYVAL with constant "true", mainly because it was quite confusing in cases where the type we were dealing with wasn't float8. I left the associated pg_control and Pg_magic_struct fields in place. Perhaps we should get rid of them, but it would save little, so it doesn't seem worth thinking hard about the compatibility implications. I just labeled them "vestigial" in places where that seemed helpful. Author: Tom Lane <tgl@sss.pgh.pa.us> Reviewed-by: Peter Eisentraut <peter@eisentraut.org> Discussion: https://postgr.es/m/1749799.1752797397@sss.pgh.pa.us Branch ------ master Details ------- https://git.postgresql.org/pg/commitdiff/ee54046601de2d14ca9107ba04c50178d9b52fe6 Modified Files -------------- contrib/btree_gist/btree_time.c | 51 ++++++++---------- contrib/btree_gist/btree_ts.c | 39 ++++++-------- src/backend/access/common/tupdesc.c | 2 +- src/backend/access/index/indexam.c | 5 -- src/backend/access/transam/xlog.c | 20 +------ src/backend/catalog/genbki.pl | 3 +- src/backend/optimizer/plan/planagg.c | 2 +- src/backend/optimizer/plan/planner.c | 4 +- src/backend/parser/parse_node.c | 2 +- src/backend/rewrite/rewriteSearchCycle.c | 2 +- src/backend/utils/adt/arrayfuncs.c | 6 +-- src/backend/utils/adt/int8.c | 75 +++++---------------------- src/backend/utils/adt/numeric.c | 74 ++++++-------------------- src/backend/utils/adt/orderedsetaggs.c | 2 +- src/backend/utils/adt/rangetypes_typanalyze.c | 2 +- src/backend/utils/fmgr/fmgr.c | 35 ------------- src/bin/initdb/initdb.c | 3 -- src/bin/pg_resetwal/pg_resetwal.c | 2 +- src/include/c.h | 8 +-- src/include/catalog/pg_type.dat | 16 +++--- src/include/fmgr.h | 2 +- src/include/postgres.h | 58 ++------------------- 22 files changed, 100 insertions(+), 313 deletions(-) ^ permalink raw reply [nested|flat] 2+ messages in thread
end of thread, other threads:[~2025-08-13 21:18 UTC | newest] Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed) -- links below jump to the message on this page -- 2025-08-13 21:18 pgsql: Grab the low-hanging fruit from forcing sizeof(Datum) to 8. Tom Lane <tgl@sss.pgh.pa.us> 2025-08-13 21:18 pgsql: Grab the low-hanging fruit from forcing USE_FLOAT8_BYVAL to true Tom Lane <tgl@sss.pgh.pa.us>
This inbox is served by DDX for PostgreSQL; see mirroring instructions for how to clone and mirror all data and code used for this inbox