On Wed, Aug 16, 2023 at 5:17 PM John Naylor
wrote:
>
> Split out tiebreaker comparisons from comparetup_* functions
Hammerkop is getting OOM errors with this change. Most grand totals in the
memory context stats show between 1 and 3 MB, and "tuplesort main" only
shows small-ish requests, so not s
Don't probe extra libraries for fdatasync.
Commit d2e15083 got rid of the main configure probe and HAVE_FDATASYNC
macro, but we still searched -lrt and -lposix4 for old Solaris systems.
It's in the C library on modern Solaris, as on other supported systems.
Reviewed-by: Peter Eisentraut
Reviewed
Remove traces of Sun -lposix4.
This was a library on ancient Solaris systems, which was eventually
replaced by -lrt, itself now redundant on that OS.
Reviewed-by: Peter Eisentraut
Reviewed-by: Tristan Partin
Discussion:
https://postgr.es/m/CA%2BhUKGLLVx4drdvXats9PxH3eeB%2BE2NkJReJ%2BRGmaOpU%3D
Invalidate smgr_targblock in smgrrelease().
In rare circumstances involving relfilenode reuse, it might have been
possible for smgr_targblock to finish up pointing past the end.
Oversight in b74e94dc. Back-patch to 15.
Reviewed-by: Heikki Linnakangas
Discussion:
https://postgr.es/m/CA%2BhUKGJ
Invalidate smgr_targblock in smgrrelease().
In rare circumstances involving relfilenode reuse, it might have been
possible for smgr_targblock to finish up pointing past the end.
Oversight in b74e94dc. Back-patch to 15.
Reviewed-by: Heikki Linnakangas
Discussion:
https://postgr.es/m/CA%2BhUKGJ
Invalidate smgr_targblock in smgrrelease().
In rare circumstances involving relfilenode reuse, it might have been
possible for smgr_targblock to finish up pointing past the end.
Oversight in b74e94dc. Back-patch to 15.
Reviewed-by: Heikki Linnakangas
Discussion:
https://postgr.es/m/CA%2BhUKGJ
doc: PG 16 relnotes, add links to doc sections
Backpatch-through: 16 only
Branch
--
REL_16_STABLE
Details
---
https://git.postgresql.org/pg/commitdiff/11f4935490eb8597fcc0e3e71c2caf90e6154943
Modified Files
--
doc/src/sgml/release-16.sgml | 354 +
sepgsql: Adjust regression expected output
Oversight in 352ea3a, where support for these subcommands has been
added.
Per buildfarm member rhinoceros.
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/d3c25ca1bf59bc4ccf589942f94bc5b945b1f46f
Modified Files
--
Add OAT hook calls for more subcommands of ALTER TABLE
The OAT hooks are added in ALTER TABLE for the following subcommands:
- { ENABLE | DISABLE | [NO] FORCE } ROW LEVEL SECURITY
- { ENABLE | DISABLE } TRIGGER
- { ENABLE | DISABLE } RULE. Note that there was hook for pg_rewrite,
but not for rela
doc: PG 16 relnotes, initial markup
Still need to add links to doc sections.
Backpatch-through: 16 only
Branch
--
REL_16_STABLE
Details
---
https://git.postgresql.org/pg/commitdiff/6b86a3e0b421a0e0c9cfa02ddc0525e431379ee4
Modified Files
--
doc/src/sgml/release-16.sgml | 54
Unify some error messages
We had essentially the same error in several different wordings.
Unify that.
Branch
--
REL_16_STABLE
Details
---
https://git.postgresql.org/pg/commitdiff/c672262779833a6b58cf5a4e85a9b76efd660cb7
Modified Files
--
src/backend/utils/adt/json.c
Unify some error messages
We had essentially the same error in several different wordings.
Unify that.
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/dca20013ebab0f0176e610a36b846d899af1d451
Modified Files
--
src/backend/utils/adt/json.c
Improved CREATE SUBSCRIPTION message for clarity
Discussion:
https://www.postgresql.org/message-id/cahut+ptfzq7jrkb0-y_uejaxalq17-bgmvv4mjjhcpop3ml...@mail.gmail.com
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/1e7ca1189cccff9851d4f12dacadf64b04fa
Modified
Improved CREATE SUBSCRIPTION message for clarity
Discussion:
https://www.postgresql.org/message-id/cahut+ptfzq7jrkb0-y_uejaxalq17-bgmvv4mjjhcpop3ml...@mail.gmail.com
Branch
--
REL_16_STABLE
Details
---
https://git.postgresql.org/pg/commitdiff/842b65905046be8a979a67484c9fd4110e4e2dfe
Mo
Remove incorrect field from information schema
The source code comment already said that the presence of the field
element_types.domain_default might be a bug in the standard, since it
never made sense there. Indeed, the field is gone in newer versions
of the standard. So just remove it.
Branch
Split out tiebreaker comparisons from comparetup_* functions
Previously, if a specialized comparator found equal datum1 keys,
the "comparetup" function would repeat the comparison on the
datum before proceeding with the unabbreviated first key
and/or additional sort keys.
Move comparing additiona
16 matches
Mail list logo