Remove dependence on -fwrapv semantics in a few places.
authorNathan Bossart <nathan@postgresql.org>
Thu, 15 Aug 2024 20:47:31 +0000 (15:47 -0500)
committerNathan Bossart <nathan@postgresql.org>
Thu, 15 Aug 2024 20:47:31 +0000 (15:47 -0500)
commit9e9a2b7031f64e49fcaf28f21a4e70eb1212165f
treee7abac509a29b705ad76d8d5123894fe3e1bf3ba
parentad89d71978429c61647ae57174a61deb192bd51c
Remove dependence on -fwrapv semantics in a few places.

This commit attempts to update a few places, such as the money,
numeric, and timestamp types, to no longer rely on signed integer
wrapping for correctness.  This is intended to move us closer
towards removing -fwrapv, which may enable some compiler
optimizations.  However, there is presently no plan to actually
remove that compiler option in the near future.

Besides using some of the existing overflow-aware routines in
int.h, this commit introduces and makes use of some new ones.
Specifically, it adds functions that accept a signed integer and
return its absolute value as an unsigned integer with the same
width (e.g., pg_abs_s64()).  It also adds functions that accept an
unsigned integer, store the result of negating that integer in a
signed integer with the same width, and return whether the negation
overflowed (e.g., pg_neg_u64_overflow()).

Finally, this commit adds a couple of tests for timestamps near
POSTGRES_EPOCH_JDATE.

Author: Joseph Koshakow
Reviewed-by: Tom Lane, Heikki Linnakangas, Jian He
Discussion: https://postgr.es/m/CAAvxfHdBPOyEGS7s%2Bxf4iaW0-cgiq25jpYdWBqQqvLtLe_t6tw%40mail.gmail.com
src/backend/utils/adt/cash.c
src/backend/utils/adt/numeric.c
src/backend/utils/adt/numutils.c
src/backend/utils/adt/timestamp.c
src/include/common/int.h
src/interfaces/ecpg/pgtypeslib/timestamp.c
src/test/regress/expected/timestamp.out
src/test/regress/expected/timestamptz.out
src/test/regress/sql/timestamp.sql
src/test/regress/sql/timestamptz.sql