Re: [Commitfest 2023-01] has started

On Sun, 8 Jan 2023 at 21:00, vignesh C <vignesh21(at)gmail(dot)com> wrote:
>
> On Tue, 3 Jan 2023 at 13:13, vignesh C <vignesh21(at)gmail(dot)com> wrote:
> >
> > Hi All,
> >
> > Just a reminder that Commitfest 2023-01 has started.
> > There are many patches based on the latest run from [1] which require
> > a) Rebased on top of head b) Fix compilation failures c) Fix test
> > failure, please have a look and rebase it so that it is easy for the
> > reviewers and committers:
> > 1. TAP output format for pg_regress
> > 2. Add BufFileRead variants with short read and EOF detection
> > 3. Add SHELL_EXIT_CODE variable to psql
> > 4. Add foreign-server health checks infrastructure
> > 5. Add last_vacuum_index_scans in pg_stat_all_tables
> > 6. Add index scan progress to pg_stat_progress_vacuum
> > 7. Add the ability to limit the amount of memory that can be allocated
> > to backends.
> > 8. Add tracking of backend memory allocated to pg_stat_activity
> > 9. CAST( ... ON DEFAULT)
> > 10. CF App: add "Returned: Needs more interest" close status
> > 11. CI and test improvements
> > 12. Cygwin cleanup
> > 13. Expand character set for ltree labels
> > 14. Fix tab completion MERGE
> > 15. Force streaming every change in logical decoding
> > 16. More scalable multixacts buffers and locking
> > 17. Move SLRU data into the regular buffer pool
> > 18. Move extraUpdatedCols out of RangeTblEntry
> > 19.New [relation] options engine
> > 20. Optimizing Node Files Support
> > 21. PGDOCS - Stats views and functions not in order?
> > 22. POC: Lock updated tuples in tuple_update() and tuple_delete()
> > 23. Parallelize correlated subqueries that execute within each worker
> > 24. Pluggable toaster
> > 25. Prefetch the next tuple's memory during seqscans
> > 26. Pulling up direct-correlated ANY_SUBLINK
> > 27. Push aggregation down to base relations and joins
> > 28. Reduce timing overhead of EXPLAIN ANALYZE using rdtsc
> > 29. Refactor relation extension, faster COPY
> > 30. Remove NEW placeholder entry from stored view query range table
> > 31. TDE key management patches
> > 32. Use AF_UNIX for tests on Windows (ie drop fallback TCP code)
> > 33. Windows filesystem support improvements
> > 34. making relfilenodes 56 bit
> > 35. postgres_fdw: commit remote (sub)transactions in parallel during pre-commit
> > 36.recovery modules
> >
> > Commitfest status as of now:
> > Needs review: 177
> > Waiting on Author: 47
> > Ready for Committer: 20
> > Committed: 31
> > Withdrawn: 4
> > Rejected: 0
> > Returned with Feedback: 0
> > Total: 279
> >
> > We will be needing more members to actively review the patches to get
> > more patches to the committed state. I would like to remind you that
> > each patch submitter is expected to review at least one patch from
> > another submitter during the CommitFest, those members who have not
> > picked up patch for review please pick someone else's patch to review
> > as soon as you can.
> > I'll send out reminders this week to get your patches rebased and
> > update the status of the patch accordingly.
> >
> > [1] - http://cfbot.cputube.org/
>
> Hi Hackers,
>
> Here's a quick status report after the first week (I think only about
> 9 commits happened during the week, the rest were pre-CF activity):
>
> status | 3rd Jan | w1
> -------------------------+-----------+-----
> Needs review: | 177 | 149
> Waiting on Author: | 47 | 60
> Ready for Committer: | 20 | 23
> Committed: | 31 | 40
> Withdrawn: | 4 | 7
> Rejected: | 0 | 0
> Returned with Feedback: | 0 | 0
> Total: | 279 | 279
>
> Here is a list of "Needs review" entries for which there has not been
> much communication on the thread and needs help in proceeding further.
> Please pick one of these and help us on how to proceed further:
> pgbench: using prepared BEGIN statement in a pipeline could cause an
> error | Yugo Nagata
> Fix dsa_free() to re-bin segment | Dongming Liu
> pg_rewind: warn when checkpoint hasn't happened after promotion | James Coleman
> Work around non-atomic read of read of control file on ext4 | Thomas Munro
> Rethinking the implementation of ts_headline | Tom Lane
> Fix GetWALAvailability function code comments for WALAVAIL_REMOVED
> return value | sirisha chamarti
> Function to log backtrace of postgres processes | vignesh C, Bharath Rupireddy
> disallow HEAP_XMAX_COMMITTED and HEAP_XMAX_IS_LOCKED_ONLY | Nathan Bossart
> New hooks in the connection path | Bertrand Drouvot
> Check consistency of GUC defaults between .sample.conf and
> pg_settings.boot_val | Justin Pryzby
> Add <<none>> support to sepgsql_restorecon | Joe Conway
> pg_stat_statements and "IN" conditions | Dmitry Dolgov
> Patch to implement missing join selectivity estimation for range types
> | Zhicheng Luo, Maxime Schoemans, Diogo Repas, Mahmoud SAKR
> Operation log for major operations | Dmitry Koval
> Consider parallel for LATERAL subqueries having LIMIT/OFFSET | James Coleman
> Using each rel as both outer and inner for anti-joins | Richard Guo
> partIndexlist for partitioned tables uniqueness | Arne Roland
> In-place persistence change of a relation (fast ALTER TABLE ... SET
> LOGGED with wal_level=minimal) | Kyotaro Horiguchi
> Speed up releasing of locks | Andres Freund, David Rowley
> nbtree performance improvements through specialization on key shape |
> Matthias van de Meent
> Add sortsupport for range types and btree_gist | Christoph Heiss
> asynchronous execution support for Custom Scan | KaiGai Kohei, kazutaka onishi
>
> Here is a list of "Ready for Committer" entries for which there has
> not been much communication on the thread and needs help in proceeding
> further. If any of the committers has some time to spare, please help
> us on these:
> Fix assertion failure with barriers in parallel hash join | Thomas
> Munro, Melanie Plageman
> pg_dump - read data for some options from external file | Pavel Stehule
> Add non-blocking version of PQcancel | Jelte Fennema
> reduce impact of lengthy startup and checkpoint tasks | Nathan Bossart
> pg_stat_statements: Track statement entry timestamp | Andrei Zubkov
> Add Amcheck option for checking unique constraints in btree indexes |
> Maxim Orlov, Pavel Borisov, Anastasia Lubennikova
> Introduce a new view for checkpointer related stats | Bharath Rupireddy
> Parallel Hash Full Join | Melanie Plageman
> Use fadvise in wal replay | Kirill Reshke, Jakub Wartak
> pg_receivewal fail to streams when the partial file to write is not
> fully initialized present in the wal receiver directory | Bharath
> Rupireddy, SATYANARAYANA NARLAPURAM
> Let libpq reject unexpected authentication requests | Jacob Champion
> Support % wildcard in extension upgrade scripts | Sandro Santilli
> TAP output format for pg_regress | Daniel Gustafsson
>
> If you have submitted a patch and it's in "Waiting for author" state,
> please aim to get it to "Needs review" state soon if you can, as
> that's where people are most likely to be looking for things to
> review.
>
> I have pinged most threads that are in "Needs review" state and don't
> apply, compile warning-free, or pass check-world. I'll do some more
> of that sort of thing, and I'll highlight a different set of patches
> next week.

Hi Hackers,

Here's a quick status report after the second week, there has been 13
entries which were committed in the last week:

status | 3rd Jan | w1 | w2
-------------------------+-----------+-------+-----
Needs review: | 177 | 149 | 128
Waiting on Author: | 47 | 60 | 64
Ready for Committer: | 20 | 23 | 26
Committed: | 31 | 40 | 53
Withdrawn: | 4 | 7 | 7
Rejected: | 0 | 0 | 0
Returned with Feedback: | 0 | 0 | 1
Total: | 279 | 279 | 279

Here is a few different patches which "Needs review", please pick one
of these and help us in proceeding further:
1) Add semi-join pushdown to postgres_fdw | Alexander Pyhalov
2) pg_upgrade test failure | Thomas Munro
3) Fix progress report of CREATE INDEX for nested partitioned tables |
Ilya Gladyshev
4) Non-replayable WAL records through overflows and >MaxAllocSize
lengths | Matthias van de Meent
5) Add sslmode "no-clientcert" to avoid auth failure in md5/scram
connections | Jim Jones
6) Add SHELL_EXIT_CODE variable to psql | Corey Huinker
7) Fix GetWALAvailability function code comments for WALAVAIL_REMOVED
return value | sirisha chamarthi
8) New strategies for freezing, advancing relfrozenxid early | Peter Geoghegan
9) Lockless queue of waiters based on atomic operations for LWLock |
Alexander Korotkov, Pavel Borisov
10) Refactor relation extension, faster COPY | Andres Freund
11) Add system view tracking shared buffer actions | Melanie Plageman
12) Add index scan progress to pg_stat_progress_vacuum | Sami Imseih
13) HOT chain validation in verify_heapam() | Himanshu Upadhyaya
14) Periodic burst growth of the checkpoint_req counter on replica. |
Anton Melnikov
15) Add EXPLAIN option GENERIC_PLAN for parameterized queries | Laurenz Albe
16) More scalable multixacts buffers and locking | Kyotaro Horiguchi ,
Andrey Borodin , Ivan Lazarev
17) In-place persistence change of a relation (fast ALTER TABLE ...
SET LOGGED with wal_level=minimal) | Kyotaro Horiguchi
18) Reducing planning time when tables have many partitions | Yuya Watari
19) ALTER TABLE and CLUSTER fail to use a BulkInsertState for toast
tables | Justin Pryzby
20) Reduce timing overhead of EXPLAIN ANALYZE using rdtsc | Andres
Freund, Lukas Fittl, David Geier

Here is a few different patches which are in "Ready for Committer"
state, if any of the committers has some time to spare, please have a
look:
1) Support load balancing in libpq | Jelte Fennema
2) Use the system CA pool for certificate verification |Jacob
Champion, Thomas Habets
3) PG DOCS - pub/sub - specifying optional parameters without values.
| Peter Smith
4) Doc: Rework contrib appendix -- informative titles, tweaked
sentences | Karl Pinc
5) Amcheck verification of GiST and GIN | Andrey Borodin, Heikki
Linnakangas, Grigory Kryachko
6) Introduce a new view for checkpointer related stats | Bharath Rupireddy
7) Faster pglz compression | Andrey Borodin, tinsane
8) AcquireExecutorLocks() and run-time pruning | Amit Langote
9) Parallel Aggregates for string_agg and array_agg | David Rowley
10) Simplify standby state machine a bit in
WaitForWALToBecomeAvailable() | Bharath Rupireddy

If you have submitted a patch and it's in "Waiting for author" state,
please aim to get it to "Needs review" state soon if you can, as
that's where people are most likely to be looking for things to
review.

I have pinged most threads that are in "Needs review" state and don't
apply, compile warning-free, or pass check-world. I'll do some more
of that sort of thing, and I'll highlight a different set of patches
next week.

Regards,
Vignesh