[COMMITTERS] pgsql: Coordinate log_line_prefix options 'm' and 'n' to share a timeva

Coordinate log_line_prefix options 'm' and 'n' to share a timeval. Commit f828654e introduced the 'n' option, but it invoked gettimeofday() independently of the 'm' option. If both options were in use (or multiple 'n' options), or if 'n' was in use along with csvlog, then the reported times could

[COMMITTERS] pgsql: Add log_line_prefix option 'n' for Unix epoch.

Add log_line_prefix option 'n' for Unix epoch. Prints time as Unix epoch with milliseconds. Tomas Vondra, reviewed by Fabien Coelho. Branch -- master Details --- http://git.postgresql.org/pg/commitdiff/f828654e1097bdbb40fedebd9ac1ea045b183e15 Modified Files -- doc/src/sgml/

Re: [COMMITTERS] pgsql: Invent a memory context reset/delete callback mechanism.

ion so that it doesn't need to be done for every input tuple of HashAgg. Thoughts? Regards, Jeff Davis -- Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-committers

[COMMITTERS] pgsql: Rename variable in AllocSetContextCreate to be consistent.

Rename variable in AllocSetContextCreate to be consistent. Everywhere else in the file, "context" is of type MemoryContext and "set" is of type AllocSet. AllocSetContextCreate uses a variable of type AllocSet, so rename it from "context" to "set". Branch -- master Details --- http://git.

[COMMITTERS] pgsql: In array_agg(), don't create a new context for every group.

In array_agg(), don't create a new context for every group. Previously, each new array created a new memory context that started out at 8kB. This is incredibly wasteful when there are lots of small groups of just a few elements each. Change initArrayResult() and friends to accept a "subcontext" a

[COMMITTERS] pgsql: Allow multibyte characters as escape in SIMILAR TO and SUBSTRING

Allow multibyte characters as escape in SIMILAR TO and SUBSTRING. Previously, only a single-byte character was allowed as an escape. This patch allows it to be a multi-byte character, though it still must be a single character. Reviewed by Heikki Linnakangas and Tom Lane. Branch -- master D

Re: [COMMITTERS] pgsql: Do all-visible handling in lazy_vacuum_page() outside its critic

g a HEAP2_VISIBLE there is that removing the tuples in the _CLEAN > step increases the chance of it the page now being all visible? It's > separate steps. When setting all-visible was part of the same critical section doing the logging for HEAP2_CLEAN, it seemed to be a par

Re: [COMMITTERS] pgsql: Do all-visible handling in lazy_vacuum_page() outside its critic

_CLEAN action, and it would look a lot more like setting a tuple hint. Also, we might be able to simplify visibilitymap_set if the logging of the heap page for checksums could be done beforehand with this variant of MarkBufferDirtyHint. Regards, Jeff Davis [1] Not that we didn't make

Re: [COMMITTERS] pgsql: Fix documentation template for CREATE TRIGGER.

rred deferrable for each row execute procedure dummy_trigger(); That case may be a bug; and regardless I don't think we want to document/encourage specifying them in that order. But the first case seems reasonable to me. Regards, Jeff Davis -- Sent via pgsql-committers m

Re: [COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

On Tue, 2014-05-06 at 23:26 -0400, Tom Lane wrote: > Jeff Davis writes: > > Fix interval test, which was broken for floating-point timestamps. > > Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that > > couldn't be made consistent between

[COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

Fix interval test, which was broken for floating-point timestamps. Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that couldn't be made consistent between integer and floating-point timestamps. It was designed to test the longest possible interval output length, so removing fou

[COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

Fix interval test, which was broken for floating-point timestamps. Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that couldn't be made consistent between integer and floating-point timestamps. It was designed to test the longest possible interval output length, so removing fou

[COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

Fix interval test, which was broken for floating-point timestamps. Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that couldn't be made consistent between integer and floating-point timestamps. It was designed to test the longest possible interval output length, so removing fou

[COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

Fix interval test, which was broken for floating-point timestamps. Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that couldn't be made consistent between integer and floating-point timestamps. It was designed to test the longest possible interval output length, so removing fou

[COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

Fix interval test, which was broken for floating-point timestamps. Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that couldn't be made consistent between integer and floating-point timestamps. It was designed to test the longest possible interval output length, so removing fou

[COMMITTERS] pgsql: Fix interval test, which was broken for floating-point timestamp

Fix interval test, which was broken for floating-point timestamps. Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that couldn't be made consistent between integer and floating-point timestamps. It was designed to test the longest possible interval output length, so removing fou

[COMMITTERS] pgsql: Improve comment for tricky aspect of index-only scans.

Improve comment for tricky aspect of index-only scans. Index-only scans avoid taking a lock on the VM buffer, which would cause a lot of contention. To be correct, that requires some intricate assumptions that weren't completely documented in the previous comment. Reviewed by Robert Haas. Branch

[COMMITTERS] pgsql: Fix crash in json_to_record().

Fix crash in json_to_record(). json_to_record() depends on get_call_result_type() for the tuple descriptor of the record that should be returned, but in some cases that cannot be determined. Add a guard to check if the tuple descriptor has been properly resolved, similar to other callers of get_ca

Re: [COMMITTERS] pgsql: Introduce replication slots.

Also, in 25.2.6: "or by storing the segments in an archive using restore_command". I think you meant "archive_command", right? Jeff Davis -- Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org) To make changes to your subscription: http:

Re: [COMMITTERS] pgsql: arrays: tighten checks for multi-dimensional input

the original semantics intentional? I didn't immediately find the thread where this was discussed, so if you send a link that might answer my question. Regards, Jeff Davis -- Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org) To make changes to your subscriptio

[COMMITTERS] pgsql: Add missing entry for session_preload_libraries in sample config

Add missing entry for session_preload_libraries in sample config. The omission was apparently an oversight in the original patch. Branch -- master Details --- http://git.postgresql.org/pg/commitdiff/7cc0ba9f17fdae610187411439e8c01b26da15ac Modified Files -- src/backend/utils

[COMMITTERS] pgsql: Lessen library-loading log level.

Lessen library-loading log level. Previously, messages were emitted at the LOG level every time a backend preloaded a library. That was acceptable (though unnecessary) for shared_preload_libraries; but it was excessive for local_preload_libraries and session_preload_libraries. Reduce to DEBUG1. A

[COMMITTERS] pgsql: Improve Range Types and Exclusion Constraints example.

Improve Range Types and Exclusion Constraints example. Make the examples self-contained to avoid confusion. Per bug report 8367 from KOIZUMI Satoru. Branch -- REL9_2_STABLE Details --- http://git.postgresql.org/pg/commitdiff/1bb4b62d3e5209dae1f1e441b7f01a053b7587f4 Modified Files --

[COMMITTERS] pgsql: Improve Range Types and Exclusion Constraints example.

Improve Range Types and Exclusion Constraints example. Make the examples self-contained to avoid confusion. Per bug report 8367 from KOIZUMI Satoru. Branch -- REL9_3_STABLE Details --- http://git.postgresql.org/pg/commitdiff/fb843b2679d959a63f48c2f4ad52243053e1a5d4 Modified Files --

[COMMITTERS] pgsql: Revert WAL posix_fallocate() patches.

Revert WAL posix_fallocate() patches. This reverts commit 269e780822abb2e44189afaccd6b0ee7aefa7ddd and commit 5b571bb8c8d2bea610e01ae1ee7bc05adcfff528. Unfortunately, the initial patch had insufficient performance testing, and resulted in a regression. Per report by Thom Brown. Branch -- ma

[COMMITTERS] pgsql: Improve Range Types and Exclusion Constraints example.

Improve Range Types and Exclusion Constraints example. Make the examples self-contained to avoid confusion. Per bug report 8367 from KOIZUMI Satoru. Branch -- master Details --- http://git.postgresql.org/pg/commitdiff/be6fcb671e4291654391fba475f4b555d2da74f3 Modified Files -

[COMMITTERS] pgsql: Handle posix_fallocate() errors.

Handle posix_fallocate() errors. On some platforms, posix_fallocate() is available but may still return EINVAL if the underlying filesystem does not support it. So, in case of an error, fall through to the alternate implementation that just writes zeros. Per buildfarm failure and analysis by Tom

[COMMITTERS] pgsql: Use posix_fallocate() for new WAL files, where available.

Use posix_fallocate() for new WAL files, where available. This function is more efficient than actually writing out zeroes to the new file, per microbenchmarks by Jon Nelson. Also, it may reduce the likelihood of WAL file fragmentation. Jon Nelson, with review by Andres Freund, Greg Smith and me.

[COMMITTERS] pgsql: Add buffer_std flag to MarkBufferDirtyHint().

Add buffer_std flag to MarkBufferDirtyHint(). MarkBufferDirtyHint() writes WAL, and should know if it's got a standard buffer or not. Currently, the only callers where buffer_std is false are related to the FSM. In passing, rename XLOG_HINT to XLOG_FPI, which is more descriptive. Back-patch to 9

[COMMITTERS] pgsql: Add buffer_std flag to MarkBufferDirtyHint().

Add buffer_std flag to MarkBufferDirtyHint(). MarkBufferDirtyHint() writes WAL, and should know if it's got a standard buffer or not. Currently, the only callers where buffer_std is false are related to the FSM. In passing, rename XLOG_HINT to XLOG_FPI, which is more descriptive. Back-patch to 9