| Age | Commit message (Collapse) | Author |
|---|
|
The idea is to encourage more the use of these new routines across the
tree, as these offer stronger type safety guarantees than palloc().
This batch of changes includes most of the trivial changes suggested by
the author for src/backend/.
A total of 334 files are updated here. Among these files, 48 of them
have their build change slightly; these are caused by line number
changes as the new allocation formulas are simpler, shaving around 100
lines of code in total.
Similar work has been done in 0c3c5c3b06a3 and 31d3847a37be.
Author: David Geier <geidav.pg@gmail.com>
Discussion: https://postgr.es/m/ad0748d4-3080-436e-b0bc-ac8f86a3466a@gmail.com
|
|
Now that commit 06edbed47862 has introduced XLogRecPtrIsValid(), we can
use that instead of:
- XLogRecPtrIsInvalid()
- direct comparisons with InvalidXLogRecPtr
- direct comparisons with literal 0
This makes the code more consistent.
Author: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Discussion: https://postgr.es/m/aQB7EvGqrbZXrMlg@ip-10-97-1-34.eu-west-3.compute.internal
|
|
5983a4cff added CompactAttribute for storing commonly used fields from
FormData_pg_attribute. 5983a4cff didn't go to the trouble of adjusting
every location where we can use CompactAttribute rather than
FormData_pg_attribute, so here we change the remaining ones.
There are some locations where I've left the code using
FormData_pg_attribute. These are mostly in the ALTER TABLE code. Using
CompactAttribute here seems more risky as often the TupleDesc is being
changed and those changes may not have been flushed to the
CompactAttribute yet.
I've also left record_recv(), record_send(), record_cmp(), record_eq()
and record_image_eq() alone as it's not clear to me that accessing the
CompactAttribute is a win here due to the FormData_pg_attribute still
having to be accessed for most cases. Switching the relevant parts to
use CompactAttribute would result in having to access both for common
cases. Careful benchmarking may reveal that something can be done to
make this better, but in absence of that, the safer option is to leave
these alone.
In ReorderBufferToastReplace(), there was a check to skip attnums < 0
while looping over the TupleDesc. Doing this is redundant since
TupleDescs don't store < 0 attnums. Removing that code allows us to
move to using CompactAttribute.
The change in validateDomainCheckConstraint() just moves fetching the
FormData_pg_attribute into the ERROR path, which is cold due to calling
errstart_cold() and results in code being moved out of the common path.
Author: David Rowley <dgrowleyml@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/CAApHDvrMy90o1Lgkt31F82tcSuwRFHq3vyGewSRN=-QuSEEvyQ@mail.gmail.com
|
|
This commit introduces a new column mem_exceeded_count to the
pg_stat_replication_slots view. This counter tracks how often the
memory used by logical decoding exceeds the logical_decoding_work_mem
limit. The new statistic helps users determine whether exceeding the
logical_decoding_work_mem limit is a rare occurrences or a frequent
issue, information that wasn't available through existing statistics.
Bumps catversion.
Author: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: shveta malik <shveta.malik@gmail.com>
Reviewed-by: Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>
Reviewed-by: Chao Li <li.evan.chao@gmail.com>
Discussion: https://postgr.es/m/978D21E8-9D3B-40EA-A4B1-F87BABE7868C@yesql.se
|
|
Make some use of anonymous unions, which are allowed as of C11, as
examples and encouragement for future code, and to test compilers.
This commit changes the ReorderBufferTXN struct.
Reviewed-by: Chao Li <li.evan.chao@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/f00a9968-388e-4f8c-b5ef-5102e962d997%40eisentraut.org
|
|
Users of logical decoding can encounter an unexpected change of
CurrentResourceOwner and CurrentMemoryContext. The problem is that,
unlike other call sites of RollbackAndReleaseCurrentSubTransaction(), in
reorderbuffer.c we fail to restore the original values of these global
variables after being clobbered by subtransaction abort. This patch
saves the values prior to the call and restores them eventually.
In addition, logical.c and logicalfuncs.c had a hack to restore resource
owner, presumably because of lack of this restore. Remove that.
Instead, because the test coverage here is not very consistent, add an
Assert() to ensure that the resowner is kept identical; this would make
it easy to detect other cases of bugs were we fail to restore resowner
properly. This could be removed later.
This is arguably an old bug, but there appears to be no reason to
backpatch it and it's risky to do so, so refrain for now.
Author: Antonin Houska <ah@cybertec.at>
Reported-by: Mihail Nikalayeu <mihailnikalayeu@gmail.com>
Reviewed-by: Euler Taveira <euler@eulerto.com>
Discussion: https://postgr.es/m/119497.1756892972@localhost
|
|
In ReorderBufferProcessTXN(), used to send the data of a transaction to
an output plugin, INSERT ON CONFLICT changes (INTERNAL_SPEC_INSERT) are
delayed until a confirmation record arrives (INTERNAL_SPEC_CONFIRM),
updating the change being processed.
8c58624df462 has added an extra step after processing a change to update
the progress of the transaction, by calling the callback
update_progress_txn() based on the LSN stored in a change after a
threshold of CHANGES_THRESHOLD (100) is reached. This logic has missed
the fact that for an INSERT ON CONFLICT change the data is freed once
processed, hence update_progress_txn() could be called pointing to a LSN
value that's already been freed. This could result in random crashes,
depending on the workload.
Per discussion, this issue is fixed by reusing in update_progress_txn()
the LSN from the change processed found at the beginning of the loop,
meaning that for a INTERNAL_SPEC_CONFIRM change the progress is updated
using the LSN of the INTERNAL_SPEC_CONFIRM change, and not the LSN from
its INTERNAL_SPEC_INSERT change. This is actually more correct, as we
want to update the progress to point to the INTERNAL_SPEC_CONFIRM
change.
Masahiko Sawada has found a nice trick to reproduce the issue: hardcode
CHANGES_THRESHOLD at 1 and run test_decoding (test "ddl" being enough)
on an instance running valgrind. The bug has been analyzed by Ethan
Mertz, who also originally suggested the solution used in this patch.
Issue introduced by 8c58624df462, so backpatch down to v16.
Author: Ethan Mertz <ethan.mertz@gmail.com>
Co-authored-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Discussion: https://postgr.es/m/aIsQqDZ7x4LAQ6u1@paquier.xyz
Backpatch-through: 16
|
|
Logical replication requires reliable conflict detection to maintain data
consistency across nodes. To achieve this, we must prevent premature
removal of tuples deleted by other origins and their associated commit_ts
data by VACUUM, which could otherwise lead to incorrect conflict reporting
and resolution.
This patch introduces a mechanism to retain deleted tuples on the
subscriber during the application of concurrent transactions from remote
nodes. Retaining these tuples allows us to correctly ignore concurrent
updates to the same tuple. Without this, an UPDATE might be misinterpreted
as an INSERT during resolutions due to the absence of the original tuple.
Additionally, we ensure that origin metadata is not prematurely removed by
vacuum freeze, which is essential for detecting update_origin_differs and
delete_origin_differs conflicts.
To support this, a new replication slot named pg_conflict_detection is
created and maintained by the launcher on the subscriber. Each apply
worker tracks its own non-removable transaction ID, which the launcher
aggregates to determine the appropriate xmin for the slot, thereby
retaining necessary tuples.
Conflict information retention (deleted tuples and commit_ts) can be
enabled per subscription via the retain_conflict_info option. This is
disabled by default to avoid unnecessary overhead for configurations that
do not require conflict resolution or logging.
During upgrades, if any subscription on the old cluster has
retain_conflict_info enabled, a conflict detection slot will be created to
protect relevant tuples from deletion when the new cluster starts.
This is a foundational work to correctly detect update_deleted conflict
which will be done in a follow-up patch.
Author: Zhijie Hou <houzj.fnst@fujitsu.com>
Reviewed-by: shveta malik <shveta.malik@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Dilip Kumar <dilipbalaut@gmail.com>
Reviewed-by: Nisha Moond <nisha.moond412@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Discussion: https://postgr.es/m/OS0PR01MB5716BE80DAEB0EE2A6A5D1F5949D2@OS0PR01MB5716.jpnprd01.prod.outlook.com
|
|
A few places were accessing bh_size directly instead of via these
handy macros.
Author: Aleksander Alekseev <aleksander@timescale.com>
Discussion: https://postgr.es/m/CAJ7c6TPQMVL%2B028T4zuw9ZqL5Du9JavOLhBQLkJeK0RznYx_6w%40mail.gmail.com
|
|
logical decoding.
Commit 4909b38af0 introduced logic to distribute invalidation messages
from catalog-modifying transactions to all concurrent in-progress
transactions. However, since each transaction distributes not only its
original invalidation messages but also previously distributed
messages to other transactions, this leads to an exponential increase
in allocation request size for invalidation messages, ultimately
causing memory allocation failure.
This commit fixes this issue by tracking distributed invalidation
messages separately per decoded transaction and not redistributing
these messages to other in-progress transactions. The maximum size of
distributed invalidation messages that one transaction can store is
limited to MAX_DISTR_INVAL_MSG_PER_TXN (8MB). Once the size of the
distributed invalidation messages exceeds this threshold, we
invalidate all caches in locations where distributed invalidation
messages need to be executed.
Back-patch to all supported versions where we introduced the fix by
commit 4909b38af0.
Note that this commit adds two new fields to ReorderBufferTXN to store
the distributed transactions. This change breaks ABI compatibility in
back branches, affecting third-party extensions that depend on the
size of the ReorderBufferTXN struct, though this scenario seems
unlikely.
Additionally, it adds a new flag to the txn_flags field of
ReorderBufferTXN to indicate distributed invalidation message
overflow. This should not affect existing implementations, as it is
unlikely that third-party extensions use unused bits in the txn_flags
field.
Bug: #18938 #18942
Author: vignesh C <vignesh21@gmail.com>
Reported-by: Duncan Sands <duncan.sands@deepbluecap.com>
Reported-by: John Hutchins <john.hutchins@wicourts.gov>
Reported-by: Laurence Parry <greenreaper@hotmail.com>
Reported-by: Max Madden <maxmmadden@gmail.com>
Reported-by: Braulio Fdo Gonzalez <brauliofg@gmail.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Reviewed-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Discussion: https://postgr.es/m/680bdaf6-f7d1-4536-b580-05c2760c67c6@deepbluecap.com
Discussion: https://postgr.es/m/18942-0ab1e5ae156613ad@postgresql.org
Discussion: https://postgr.es/m/18938-57c9a1c463b68ce0@postgresql.org
Discussion: https://postgr.es/m/CAD1FGCT2sYrP_70RTuo56QTizyc+J3wJdtn2gtO3VttQFpdMZg@mail.gmail.com
Discussion: https://postgr.es/m/CANO2=B=2BT1hSYCE=nuuTnVTnjidMg0+-FfnRnqM6kd23qoygg@mail.gmail.com
Backpatch-through: 13
|
|
Data loss can happen when the DDLs like ALTER PUBLICATION ... ADD TABLE ...
or ALTER TYPE ... that don't take a strong lock on table happens
concurrently to DMLs on the tables involved in the DDL. This happens
because logical decoding doesn't distribute invalidations to concurrent
transactions and those transactions use stale cache data to decode the
changes. The problem becomes bigger because we keep using the stale cache
even after those in-progress transactions are finished and skip the
changes required to be sent to the client.
This commit fixes the issue by distributing invalidation messages from
catalog-modifying transactions to all concurrent in-progress transactions.
This allows the necessary rebuild of the catalog cache when decoding new
changes after concurrent DDL.
We observed performance regression primarily during frequent execution of
*publication DDL* statements that modify the published tables. The
regression is minor or nearly nonexistent for DDLs that do not affect the
published tables or occur infrequently, making this a worthwhile cost to
resolve a longstanding data loss issue.
An alternative approach considered was to take a strong lock on each
affected table during publication modification. However, this would only
address issues related to publication DDLs (but not the ALTER TYPE ...)
and require locking every relation in the database for publications
created as FOR ALL TABLES, which is impractical.
The bug exists in all supported branches, but we are backpatching till 14.
The fix for 13 requires somewhat bigger changes than this fix, so the fix
for that branch is still under discussion.
Reported-by: hubert depesz lubaczewski <depesz@depesz.com>
Reported-by: Tomas Vondra <tomas.vondra@enterprisedb.com>
Author: Shlok Kyal <shlok.kyal.oss@gmail.com>
Author: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Reviewed-by: Zhijie Hou <houzj.fnst@fujitsu.com>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Amit Kapila <amit.kapila16@gmail.com>
Tested-by: Benoit Lobréau <benoit.lobreau@dalibo.com>
Backpatch-through: 14
Discussion: https://postgr.es/m/de52b282-1166-1180-45a2-8d8917ca74c6@enterprisedb.com
Discussion: https://postgr.es/m/CAD21AoAenVqiMjpN-PvGHL1N9DWnHSq673bfgr6phmBUzx=kLQ@mail.gmail.com
|
|
There used to be bespoken pools for these structs to reduce the
palloc/pfree overhead, but that was ripped out a long time ago and
replaced with the generic, cheaper generational memory allocator
(commit a4ccc1cef5). The Get/Return terminology made sense with the
pools, as you "got" an object from the pool and "returned" it later,
but now it just looks weird. Rename to Alloc/Free.
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://www.postgresql.org/message-id/c9e43d2d-8e83-444f-b111-430377368989@iki.fi
|
|
For AIO, and also some other recent patches, we need the ability to call
relpath() in a critical section. Until now that was not feasible, as it
allocated memory.
The fact that relpath() allocated memory also made it awkward to use in log
messages because we had to take care to free the memory afterwards. Which we
e.g. didn't do for when zeroing out an invalid buffer.
We discussed other solutions, e.g. filling a pre-allocated buffer that's
passed to relpath(), but they all came with plenty downsides or were larger
projects. The easiest fix seems to be to make relpath() return the path by
value.
To be able to return the path by value we need to determine the maximum length
of a relation path. This patch adds a long #define that computes the exact
maximum, which is verified to be correct in a regression test.
As this change the signature of relpath(), extensions using it will need to
adapt their code. We discussed leaving a backward-compat shim in place, but
decided it's not worth it given the use of relpath() doesn't seem widespread.
Discussion: https://postgr.es/m/xeri5mla4b5syjd5a25nok5iez2kr3bm26j2qn4u7okzof2bmf@kwdh2vf7npra
|
|
RBTXN_PREPARE flag and rbtxn_prepared macro could be misinterpreted as
either indicating the transaction type (e.g. a prepared transaction or
a normal transaction) or its currentstate (e.g. skipped or its prepare
message is sent), especially after commit 072ee847ad4 introduced the
RBTXN_SENT_PREPARE flag and the rbtxn_sent_prepare macro.
The RBTXN_PREPARE flag (and its corresponding macro) have been renamed
to RBTXN_IS_PREPARE to explicitly indicate the transaction
type. Therefore, this commit also adds the RBTXN_IS_PREPARE flag to
the transaction that is a prepared transaction and has been skipped,
which previously had only the RBTXN_SKIPPED_PREPARE flag.
Reviewed-by: Amit Kapila, Peter Smith
Discussion: https://postgr.es/m/CAA4eK1KgNmBsG%3D155E7QQ6TX9RoWnM4z5Z20SvsbwxSe_QXYsg%40mail.gmail.com
|
|
Previously, transaction aborts were detected concurrently only during
system catalog scans while replaying a transaction in streaming mode.
This commit adds an additional CLOG lookup to check the transaction
status, allowing the logical decoding to skip changes also when it
doesn't touch system catalogs, if the transaction is already
aborted. This optimization enhances logical decoding performance,
especially for large transactions that have already been rolled back,
as it avoids unnecessary disk or network I/O.
To avoid potential slowdowns caused by frequent CLOG lookups for small
transactions (most of which commit), the CLOG lookup is performed only
for large transactions before eviction. The performance benchmark
results showed there is not noticeable performance regression due to
CLOG lookups.
Reviewed-by: Amit Kapila, Peter Smith, Vignesh C, Ajin Cherian
Reviewed-by: Dilip Kumar, Andres Freund
Discussion: https://postgr.es/m/CAD21AoDht9Pz_DFv_R2LqBTBbO4eGrpa9Vojmt5z5sEx3XwD7A@mail.gmail.com
|
|
Consistently use "Size" (or size_t, or in some places int64 or double)
as the type for variables holding memory allocation sizes. In most
places variables' data types were fine already, but we had an ancient
habit of computing bytes from kilobytes-units GUCs with code like
"work_mem * 1024L". That risks overflow on Win64 where they did not
make "long" as wide as "size_t". We worked around that by restricting
such GUCs' ranges, so you couldn't set work_mem et al higher than 2GB
on Win64. This patch removes that restriction, after replacing such
calculations with "work_mem * (Size) 1024" or variants of that.
It should be noted that this patch was constructed by searching
outwards from the GUCs that have MAX_KILOBYTES as upper limit.
So I can't positively guarantee there are no other places doing
memory-size arithmetic in int or long variables. I do however feel
pretty confident that increasing MAX_KILOBYTES on Win64 is safe now.
Also, nothing in our code should be dealing in multiple-gigabyte
allocations without authorization from a relevant GUC, so it seems
pretty likely that this search caught everything that could be at
risk of overflow.
Author: Vladlen Popolitov <v.popolitov@postgrespro.ru>
Co-authored-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/1a01f0-66ec2d80-3b-68487680@27595217
|
|
Some places spelled it "it's", which is short for "it is".
In passing, fix a couple other nearby grammatical errors.
Author: Jacob Brazeal <jacob.brazeal@gmail.com>
Discussion: https://postgr.es/m/CA+COZaAO8g1KJCV0T48=CkJMjAnnfTGLWOATz+2aCh40c2Nm+g@mail.gmail.com
|
|
Backpatch-through: 13
|
|
This used to say "nsubxcnt isn't decreased when subtransactions
abort", but there's no variable called nsubxcnt. Commit 8548ddc61b
changed it to "subxcnt", among other typo fixes, but that was wrong
too: the comment actually talks about txn->nsubtxns. That's the field
that's incremented but never decremented and is used for the
allocation earlier in the function.
|
|
Commit a4ccc1cef introduced the Generation Context and modified the
logical decoding process to use a Generation Context with a fixed
block size of 8MB for storing tuple data decoded during logical
decoding (i.e., rb->tup_context). Several reports have indicated that
the logical decoding process can be terminated due to
out-of-memory (OOM) situations caused by excessive memory usage in
rb->tup_context.
This issue can occur when decoding a workload involving several
concurrent transactions, including a long-running transaction that
modifies tuples. By design, the Generation Context does not free a
memory block until all chunks within that block are
released. Consequently, if tuples modified by the long-running
transaction are stored across multiple memory blocks, these blocks
remain allocated until the long-running transaction completes, leading
to substantial memory fragmentation. The memory usage during logical
decoding, tracked by rb->size, does not account for memory
fragmentation, resulting in potentially much higher memory consumption
than the value of the logical_decoding_work_mem parameter.
Various improvement strategies were discussed in the relevant
thread. This change reduces the block size of the Generation Context
used in rb->tup_context from 8MB to 8kB. This modification
significantly decreases the likelihood of substantial memory
fragmentation occurring and is relatively straightforward to
backport. Performance testing across multiple platforms has confirmed
that this change will not introduce any performance degradation that
would impact actual operation.
Backport to all supported branches.
Reported-by: Alex Richman, Michael Guissine, Avi Weinberg
Reviewed-by: Amit Kapila, Fujii Masao, David Rowley
Tested-by: Hayato Kuroda, Shlok Kyal
Discussion: https://postgr.es/m/CAD21AoBTY1LATZUmvSXEssvq07qDZufV4AF-OHh9VD2pC0VY2A%40mail.gmail.com
Backpatch-through: 12
|
|
This is similar to 2065ddf5e34c, but this time for pg_logical/ itself
and its contents, like the paths for snapshots, mappings or origin
checkpoints.
Author: Bertrand Drouvot
Reviewed-by: Ashutosh Bapat, Yugo Nagata, Michael Paquier
Discussion: https://postgr.es/m/ZryVvjqS9SnV1GPP@ip-10-97-1-34.eu-west-3.compute.internal
|
|
This commit replaces most of the hardcoded values of "pg_replslot" by a
new PG_REPLSLOT_DIR #define. This makes the style more consistent with
the existing PG_STAT_TMP_DIR, for example. More places will follow a
similar change.
Author: Bertrand Drouvot
Reviewed-by: Ashutosh Bapat, Yugo Nagata, Michael Paquier
Discussion: https://postgr.es/m/ZryVvjqS9SnV1GPP@ip-10-97-1-34.eu-west-3.compute.internal
|
|
|
|
Commit 5bec1d6bc5e changed the memory usage updates of the
ReorderBufferTXN to zero all at once by subtracting txn->size, rather
than updating it for each change. However, if TOAST reconstruction
data remained in the transaction when freeing it, there were cases
where it further subtracted the memory counter from zero, resulting in
an assertion failure.
This change calculates the memory size for each change and updates the
memory usage to precisely the amount that has been freed.
Backpatch to v17, where this was introducd.
Reviewed-by: Amit Kapila, Shlok Kyal
Discussion: https://postgr.es/m/CAD21AoAqkNUvicgKPT_dXzNoOwpPkVTg0QPPxEcWmzT0moCJ1g%40mail.gmail.com
Backpatch-through: 17
|
|
This reverts commit b840508644 and bcb14f4abc. These commits were made
for commit 5bec1d6bc5 (Improve eviction algorithm in ReorderBuffer
using max-heap for many subtransactions). However, per discussion,
commit efb8acc0d0 replaced binary heap + index with pairing heap, and
made these commits unnecessary.
Reported-by: Jeff Davis
Discussion: https://postgr.es/m/12747c15811d94efcc5cda72d6b35c80d7bf3443.camel%40j-davis.com
|
|
A pairing heap can perform the same operations as the binary heap +
index, with as good or better algorithmic complexity, and that's an
existing data structure so that we don't need to invent anything new
compared to v16. This commit makes the new binaryheap functionality
that was added in commits b840508644 and bcb14f4abc unnecessary, but
they will be reverted separately.
Remove the optimization to only build and maintain the heap when the
amount of memory used is close to the limit, becuase the bookkeeping
overhead with the pairing heap seems to be small enough that it
doesn't matter in practice.
Reported-by: Jeff Davis
Author: Heikki Linnakangas
Reviewed-by: Michael Paquier, Hayato Kuroda, Masahiko Sawada
Discussion: https://postgr.es/m/12747c15811d94efcc5cda72d6b35c80d7bf3443.camel%40j-davis.com
|
|
subtransactions.
Previously, when selecting the transaction to evict during logical
decoding, we check all transactions to find the largest
transaction. This could lead to a significant replication lag
especially in the case where there are many subtransactions.
This commit improves the eviction algorithm in ReorderBuffer using the
max-heap with transaction size as the key to efficiently find the
largest transaction.
The max-heap starts with empty. While the max-heap is empty, we don't
do anything for the max-heap when updating the memory
counter. Therefore, we get the largest transaction in O(N) time, where
N is the number of transactions including top-level transactions and
subtransactions.
We build the max-heap just before selecting the largest transactions
if the number of transactions being decoded is higher than the
threshold, MAX_HEAP_TXN_COUNT_THRESHOLD. After building the max-heap,
we also update the max-heap when updating the memory counter. The
intention is to efficiently find the largest transaction in O(1) time
instead of incurring the cost of memory counter updates (O(log
N)). Once the number of transactions got lower than the threshold, we
reset the max-heap.
The performance benchmark results showed significant speed up (more
than x30 speed up on my machine) in decoding a transaction with 100k
subtransactions, whereas there is no visible overhead in other cases.
Reviewed-by: Amit Kapila, Hayato Kuroda, Vignesh C, Ajin Cherian,
Tomas Vondra, Shubham Khanna, Peter Smith, Álvaro Herrera,
Euler Taveira
Discussion: https://postgr.es/m/CAD21AoAfKTgrBrLq96GcTv9d6k97zaQcDM-rxfKEt4GSe0qnaQ%40mail.gmail.com
|
|
Previously, binaryheap didn't support updating a key and removing a
node in an efficient way. For example, in order to remove a node from
the binaryheap, the caller had to pass the node's position within the
array that the binaryheap internally has. Removing a node from the
binaryheap is done in O(log n) but searching for the key's position is
done in O(n).
This commit adds a hash table to binaryheap in order to track the
position of each nodes in the binaryheap. That way, by using newly
added functions such as binaryheap_update_up() etc., both updating a
key and removing a node can be done in O(1) on an average and O(log n)
in worst case. This is known as the indexed binary heap. The caller
can specify to use the indexed binaryheap by passing indexed = true.
The current code does not use the new indexing logic, but it will be
used by an upcoming patch.
Reviewed-by: Vignesh C, Peter Smith, Hayato Kuroda, Ajin Cherian,
Tomas Vondra, Shubham Khanna
Discussion: https://postgr.es/m/CAD21AoDffo37RC-eUuyHJKVEr017V2YYDLyn1xF_00ofptWbkg%40mail.gmail.com
|
|
Author: Kyotaro Horiguchi
Discussion: https://postgr.es/m/20240314.132817.1496502692848380820.horikyota.ntt@gmail.com
|
|
as determined by include-what-you-use (IWYU)
While IWYU also suggests to *add* a bunch of #include's (which is its
main purpose), this patch does not do that. In some cases, a more
specific #include replaces another less specific one.
Some manual adjustments of the automatic result:
- IWYU currently doesn't know about includes that provide global
variable declarations (like -Wmissing-variable-declarations), so
those includes are being kept manually.
- All includes for port(ability) headers are being kept for now, to
play it safe.
- No changes of catalog/pg_foo.h to catalog/pg_foo_d.h, to keep the
patch from exploding in size.
Note that this patch touches just *.c files, so nothing declared in
header files changes in hidden ways.
As a small example, in src/backend/access/transam/rmgr.c, some IWYU
pragma annotations are added to handle a special case there.
Discussion: https://www.postgresql.org/message-id/flat/af837490-6b2f-46df-ba05-37ea6a6653fc%40eisentraut.org
|
|
Commit 6b80394781 introduced integer comparison functions designed
to be as efficient as possible while avoiding overflow. This
commit makes use of these functions in many of the in-tree qsort()
comparators to help ensure transitivity. Many of these comparator
functions should also see a small performance boost.
Author: Mats Kindahl
Reviewed-by: Andres Freund, Fabrízio de Royes Mello
Discussion: https://postgr.es/m/CA%2B14426g2Wa9QuUpmakwPxXFWG_1FaY0AsApkvcTBy-YfS6uaw%40mail.gmail.com
|
|
Since commit a4ccc1cef, the 'node' and 'alloc_tuple_size' fields of
the ReorderBufferTupleBuf structure are no longer used. This leaves
only the 'tuple' field in the structure. Since keeping a single-field
structure makes little sense, the ReorderBufferTupleBuf is removed
entirely. The code is refactored accordingly.
No back-patching since these are ABI changes in an exposed structure
and functions, and there would be some risk of breaking extensions.
Author: Aleksander Alekseev
Reviewed-by: Amit Kapila, Masahiko Sawada, Reid Thompson
Discussion: https://postgr.es/m/CAD21AoCvnuxiXXfRecp7g9+CeC35POQfhuQeJFr7_9u_Q5jc_Q@mail.gmail.com
|
|
Reported-by: Michael Paquier
Discussion: https://postgr.es/m/ZZKTDPxBBMt3C0J9@paquier.xyz
Backpatch-through: 12
|
|
The code is able to compile already without warnings under
-Wshadow=compatible-local, which is itself already enabled in the tree,
and the ones fixed here showed up with the more restrictive -Wshadow.
There are more of these that we may want to look at, and the ones fixed
here made the code confusing.
Author: Peter Smith
Discussion: https://postgr.es/m/CAHut+PuR0y4ofNOxi691VTVWmBfScHV9AaBMGSpeh8+DKp81Nw@mail.gmail.com
|
|
The logical_replication_mode GUC is intended for testing and debugging
purposes, but its current name may be misleading and encourage users to make
unnecessary changes.
To avoid confusion, renaming the GUC to a less misleading name
debug_logical_replication_streaming that casual users are less likely to mistakenly
assume needs to be modified in a regular logical replication setup.
Author: Hou Zhijie <houzj.fnst@cn.fujitsu.com>
Reviewed-by: Peter Smith <smithpb2250@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/d672d774-c44b-6fec-f993-793e744f169a%40eisentraut.org
|
|
Commit 7259736a6 updated the comment but it was not correct since
ReorderBufferLargestStreamableTopTXN() returns only top-level
transactions.
Reviewed-by: Amit Kapila
Discussion: https://postgr.es/m/CAD21AoA9XB7OR86BqvrCe2dMYX%2BZv3-BvVmjF%3DGY2z6jN-kqjg%40mail.gmail.com
Backpatch-through: 14
|
|
Run pgindent, pgperltidy, and reformat-dat-files.
This set of diffs is a bit larger than typical. We've updated to
pg_bsd_indent 2.1.2, which properly indents variable declarations that
have multi-line initialization expressions (the continuation lines are
now indented one tab stop). We've also updated to perltidy version
20230309 and changed some of its settings, which reduces its desire to
add whitespace to lines to make assignments etc. line up. Going
forward, that should make for fewer random-seeming changes to existing
code.
Discussion: https://postgr.es/m/20230428092545.qfb3y5wcu4cm75ur@alvherre.pgsql
|
|
The changes done in this commit impact comments with no direct
user-visible changes, with fixes for incorrect function, variable or
structure names.
Author: Alexander Lakhin
Discussion: https://postgr.es/m/e8c38840-596a-83d6-bd8d-cebc51111572@gmail.com
|
|
Currently, there are quite a few places in reorderbuffer.c that tries to
access top-transaction for a subtransaction. This makes the code to access
top-transaction consistent and easier to follow.
Author: Peter Smith
Reviewed-by: Vignesh C, Sawada Masahiko
Discussion: https://postgr.es/m/CAHut+PuCznOyTqBQwjRUu-ibG-=KHyCv-0FTcWQtZUdR88umfg@mail.gmail.com
|
|
Whe decoding a transactional logical message, logicalmsg_decode called
SnapBuildGetOrBuildSnapshot. But we may not have a consistent snapshot
yet at that point. We don't actually need the snapshot in this case
(during replay we'll have the snapshot from the transaction), so in
practice this is harmless. But in assert-enabled build this crashes.
Fixed by requesting the snapshot only in non-transactional case, where
we are guaranteed to have SNAPBUILD_CONSISTENT.
Backpatch to 11. The issue exists since 9.6.
Backpatch-through: 11
Reviewed-by: Andres Freund
Discussion: https://postgr.es/m/84d60912-6eab-9b84-5de3-41765a5449e8@enterprisedb.com
|
|
The DDLs like Refresh Materialized views that generate lots of temporary
data due to rewrite rules may not be processed by output plugins (for
example pgoutput). So, we won't send keep-alive messages for a long time
while processing such commands and that can lead the subscriber side to
timeout. We have previously fixed a similar case for large transactions in
commit f95d53eded where the output plugin filters all or most of the
changes but missed to handle the DDLs.
We decided not to backpatch this as this adds a new callback in the
existing exposed structure and moreover, users can increase the
wal_sender_timeout and wal_receiver_timeout to avoid this problem.
Author: Wang wei, Hou Zhijie
Reviewed-by: Peter Smith, Ashutosh Bapat, Shi yu, Amit Kapila
Discussion: https://postgr.es/m/OS3PR01MB6275478E5D29E4A563302D3D9E2B9@OS3PR01MB6275.jpnprd01.prod.outlook.com
Discussion: https://postgr.es/m/CAA5-nLARN7-3SLU_QUxfy510pmrYK6JJb=bk3hcgemAM_pAv+w@mail.gmail.com
|
|
Some of these appear to be leftovers from when hash_search() took a
char * argument (changed in 5999e78fc45dcb91784b64b6e9ae43f4e4f68ca2).
Since after this there is some more horizontal space available, do
some light reformatting where suitable.
Reviewed-by: Corey Huinker <corey.huinker@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/fd9adf5d-b1aa-e82f-e4c7-263c30145807%40enterprisedb.com
|
|
Rename the developer option 'logical_decoding_mode' to the more flexible
name 'logical_replication_mode' because doing so will make it easier to
extend this option in the future to help test other areas of logical
replication.
Currently, it is used on the publisher side to allow streaming or
serializing each change in logical decoding. In the upcoming patch, we are
planning to use it on the subscriber. On the subscriber, it will allow
serializing the changes to file and notifies the parallel apply workers to
read and apply them at the end of the transaction.
We discussed exposing this parameter as a subscription option but
it did not seem advisable since it is primarily used for testing/debugging
and there is no other such parameter. We also discussed having separate
GUCs for publisher and subscriber but for current testing/debugging
requirements, one GUC is sufficient.
Author: Hou Zhijie
Reviewed-by: Peter Smith, Kuroda Hayato, Sawada Masahiko, Amit Kapila
Discussion: https://postgr.es/m/CAD21AoAy2c=Mx=FTCs+EwUsf2kQL5MmU3N18X84k0EmCXntK4g@mail.gmail.com
Discussion: https://postgr.es/m/CAA4eK1+wyN6zpaHUkCLorEWNx75MG0xhMwcFhvjqm2KURZEAGw@mail.gmail.com
|
|
Currently, for large transactions, the publisher sends the data in
multiple streams (changes divided into chunks depending upon
logical_decoding_work_mem), and then on the subscriber-side, the apply
worker writes the changes into temporary files and once it receives the
commit, it reads from those files and applies the entire transaction. To
improve the performance of such transactions, we can instead allow them to
be applied via parallel workers.
In this approach, we assign a new parallel apply worker (if available) as
soon as the xact's first stream is received and the leader apply worker
will send changes to this new worker via shared memory. The parallel apply
worker will directly apply the change instead of writing it to temporary
files. However, if the leader apply worker times out while attempting to
send a message to the parallel apply worker, it will switch to
"partial serialize" mode - in this mode, the leader serializes all
remaining changes to a file and notifies the parallel apply workers to
read and apply them at the end of the transaction. We use a non-blocking
way to send the messages from the leader apply worker to the parallel
apply to avoid deadlocks. We keep this parallel apply assigned till the
transaction commit is received and also wait for the worker to finish at
commit. This preserves commit ordering and avoid writing to and reading
from files in most cases. We still need to spill if there is no worker
available.
This patch also extends the SUBSCRIPTION 'streaming' parameter so that the
user can control whether to apply the streaming transaction in a parallel
apply worker or spill the change to disk. The user can set the streaming
parameter to 'on/off', or 'parallel'. The parameter value 'parallel' means
the streaming will be applied via a parallel apply worker, if available.
The parameter value 'on' means the streaming transaction will be spilled
to disk. The default value is 'off' (same as current behaviour).
In addition, the patch extends the logical replication STREAM_ABORT
message so that abort_lsn and abort_time can also be sent which can be
used to update the replication origin in parallel apply worker when the
streaming transaction is aborted. Because this message extension is needed
to support parallel streaming, parallel streaming is not supported for
publications on servers < PG16.
Author: Hou Zhijie, Wang wei, Amit Kapila with design inputs from Sawada Masahiko
Reviewed-by: Sawada Masahiko, Peter Smith, Dilip Kumar, Shi yu, Kuroda Hayato, Shveta Mallik
Discussion: https://postgr.es/m/CAA4eK1+wyN6zpaHUkCLorEWNx75MG0xhMwcFhvjqm2KURZEAGw@mail.gmail.com
|
|
After restart, we try to stream the changes for large transactions that
were not sent before server crash and restart. However, we forget to send
the abort message for such transactions. This leads to spurious streaming
files on the subscriber which won't be cleaned till the apply worker or
the subscriber server restarts.
Reported-by: Dilip Kumar
Author: Hou Zhijie
Reviewed-by: Dilip Kumar and Amit Kapila
Backpatch-through: 14
Discussion: https://postgr.es/m/OS0PR01MB5716A773F46768A1B75BE24394FB9@OS0PR01MB5716.jpnprd01.prod.outlook.com
|
|
Backpatch-through: 11
|
|
This enables streaming or serializing changes immediately in logical
decoding. This parameter is intended to be used to test logical decoding
and replication of large transactions for which otherwise we need to
generate the changes till logical_decoding_work_mem is reached.
This helps in reducing the timing of existing tests related to logical
replication of in-progress transactions and will help in writing tests for
for the upcoming feature for parallelly applying large in-progress
transactions.
Author: Shi yu
Reviewed-by: Sawada Masahiko, Shveta Mallik, Amit Kapila, Dilip Kumar, Kuroda Hayato, Kyotaro Horiguchi
Discussion: https://postgr.es/m/OSZPR01MB63104E7449DBE41932DB19F1FD1B9@OSZPR01MB6310.jpnprd01.prod.outlook.com
|
|
After restart, we don't perform streaming of an in-progress transaction if
it was previously decoded and confirmed by the client. To achieve that we
were comparing the END location of the WAL record being decoded with the
WAL location we have already decoded and confirmed by the client. While
decoding the commit record, to decide whether to process and send the
complete transaction, we compare its START location with the WAL location
we have already decoded and confirmed by the client. Now, if we need to
queue some change in the transaction while decoding the commit record
(e.g. snapshot), it is possible that we decide to stream the transaction
but later commit processing decides to skip it. In such a case, we would
needlessly send the changes and later when we decide to skip it, we will
send stream abort.
We also sometimes decide to stream the changes when we actually just need
to process them locally like a change for invalidations. This will lead us
to send empty streams. To avoid this, while queuing each change for
decoding, we remember whether the transaction has any change that actually
needs to be sent downstream and use that information later to decide
whether to stream the transaction or not.
Note, we can't avoid all cases where we have to send empty streams like
the case where the plugin later decides that the change is not
publishable. However, we will no longer need to send stream_abort when we
skip sending a particular transaction.
Author: Dilip Kumar
Reviewed-by: Hou Zhijie, Ashutosh Bapat, Shi yu, Amit Kapila
Discussion: https://postgr.es/m/CAFiTN-tHK=7LzfrPs8fbT2ksrOJGQbzywcgXst2bM9-rJJAAUg@mail.gmail.com
|
|
The comments atop seem to indicate that we always accumulate invalidation
messages in a top-level transaction which is neither required nor matches
with the code.
Author: Amit Kapila
Reviewd by: Masahiko Sawada
Backpatch-through: 14, where it was introduced in commit c55040ccd0
Discussion: https://postgr.es/m/CAA4eK1LxGgnUroPz8STb6OfjVU1yaHoSA+T63URwmGCLdMJ0LA@mail.gmail.com
|
|
We have various requirements when using a dlist_head to keep track of the
number of items in the list. This, traditionally, has been done by
maintaining a counter variable in the calling code. Here we tidy this up
by adding "dclist", which is very similar to dlist but also keeps track of
the number of items stored in the list.
Callers may use the new dclist_count() function when they need to know how
many items are stored. Obtaining the count is an O(1) operation.
For simplicity reasons, dclist and dlist both use dlist_node as their node
type and dlist_iter/dlist_mutable_iter as their iterator type. dclists
have all of the same functionality as dlists except there is no function
named dclist_delete(). To remove an item from a list dclist_delete_from()
must be used. This requires knowing which dclist the given item is stored
in.
Additionally, here we also convert some dlists where additional code
exists to keep track of the number of items stored and to make these use
dclists instead.
Author: David Rowley
Reviewed-by: Bharath Rupireddy, Aleksander Alekseev
Discussion: https://postgr.es/m/CAApHDvrtVxr+FXEX0VbViCFKDGxA3tWDgw9oFewNXCJMmwLjLg@mail.gmail.com
|
