Enable parallel query with SERIALIZABLE isolation.
Previously, the SERIALIZABLE isolation level prevented parallel query
from being used. Allow the two features to be used together by
sharing the leader's SERIALIZABLEXACT with parallel workers.
An extra per-SERIALIZABLEXACT LWLock is introduced
During pg_upgrade, conditionally skip transfer of FSMs.
If a heap on the old cluster has 4 pages or fewer, and the old cluster
was PG v11 or earlier, don't copy or link the FSM. This will shrink
space usage for installations with large numbers of small tables.
This will allow pg_upgrade to take a
On Thu, Mar 14, 2019 at 08:39:10AM -0400, Robert Haas wrote:
> On Thu, Mar 14, 2019 at 1:16 AM Michael Paquier wrote:
> > Fix thinko when bumping on temporary directories in pg_verify_checksums
>
> What the heck does it mean to bump on a directory?
My apologies for the confusion. It looks like
Reorder identity regression test
The previous test order had the effect that if something was wrong
with the identity functionality, the create_table_like test would
likely fail or crash first, which is confusing. Reorder so that the
identity test comes before create_table_like.
Branch
--
ma
Fix some oversights in commit 2455ab488.
The idea was to generate all the junk in a destroyable subcontext rather
than leaking it in the caller's context, but partition_bounds_create was
still being called in the caller's context, allowing plenty of scope for
leakage. Also, get_rel_relkind() was
Improve code comment
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/61dc407893600f551dbcbc235d1dccd134f43da0
Modified Files
--
src/backend/commands/copy.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
Remove unused #include
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/8bee36708f6ee4bf93be68f9a368379b0e9c6474
Modified Files
--
src/backend/executor/execMain.c | 1 -
1 file changed, 1 deletion(-)
Add BKI_DEFAULT to pg_class.relrewrite
This column is always 0 on disk, so it doesn't have to be tracked
separately for each entry.
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/b13a913607b9b304d16029361b1b17e10783cf7c
Modified Files
--
src/include/ca
Ensure dummy paths have correct required_outer if rel is parameterized.
The assertions added by commits 34ea1ab7f et al found another problem:
set_dummy_rel_pathlist and mark_dummy_rel were failing to label
the dummy paths they create with the correct outer_relids, in case
the relation is necessar
Ensure dummy paths have correct required_outer if rel is parameterized.
The assertions added by commits 34ea1ab7f et al found another problem:
set_dummy_rel_pathlist and mark_dummy_rel were failing to label
the dummy paths they create with the correct outer_relids, in case
the relation is necessar
Ensure dummy paths have correct required_outer if rel is parameterized.
The assertions added by commits 34ea1ab7f et al found another problem:
set_dummy_rel_pathlist and mark_dummy_rel were failing to label
the dummy paths they create with the correct outer_relids, in case
the relation is necessar
Ensure dummy paths have correct required_outer if rel is parameterized.
The assertions added by commits 34ea1ab7f et al found another problem:
set_dummy_rel_pathlist and mark_dummy_rel were failing to label
the dummy paths they create with the correct outer_relids, in case
the relation is necessar
Ensure dummy paths have correct required_outer if rel is parameterized.
The assertions added by commits 34ea1ab7f et al found another problem:
set_dummy_rel_pathlist and mark_dummy_rel were failing to label
the dummy paths they create with the correct outer_relids, in case
the relation is necessar
Ensure dummy paths have correct required_outer if rel is parameterized.
The assertions added by commits 34ea1ab7f et al found another problem:
set_dummy_rel_pathlist and mark_dummy_rel were failing to label
the dummy paths they create with the correct outer_relids, in case
the relation is necessar
Defend against leaks into RelationBuildPartitionDesc.
In normal builds, this isn't very important, because the leaks go
into fairly short-lived contexts, but under CLOBBER_CACHE_ALWAYS,
this can result in leaking hundreds of megabytes into MessageContext,
which probably explains recent failures on
On Thu, Mar 14, 2019 at 1:16 AM Michael Paquier wrote:
> Fix thinko when bumping on temporary directories in pg_verify_checksums
What the heck does it mean to bump on a directory?
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
Refactor ParamListInfo initialization
There were six copies of identical nontrivial code. Put it into a
function.
Branch
--
master
Details
---
https://git.postgresql.org/pg/commitdiff/c6ff0b892c30122b75d32c524109d16ee3c973f0
Modified Files
--
src/backend/commands/prepare.c
Fix volatile vs. pointer confusion
Variables used after a longjmp() need to be declared volatile. In
case of a pointer, it's the pointer itself that needs to be declared
volatile, not the pointed-to value. So we need
PyObject *volatile items;
instead of
volatile PyObject *items; /* w
18 matches
Mail list logo
