Re: Commit fest 2022-11

2022年11月4日(金) 10:23 Ian Lawrence Barwick <barwick(at)gmail(dot)com>:
>
> 2022年11月4日(金) 9:43 Justin Pryzby <pryzby(at)telsasoft(dot)com>:
> >
> > On Thu, Nov 03, 2022 at 11:33:57AM +0900, Ian Lawrence Barwick wrote:
> > > 2022年11月2日(水) 19:10 Greg Stark <stark(at)mit(dot)edu>:
> > > >
> > > > On Tue, 1 Nov 2022 at 06:56, Michael Paquier <michael(at)paquier(dot)xyz>
wrote:
> > > >
> > > > > Two people showing up to help is really great, thanks! I'll be
around
> > > > > as well this month, so I'll do my share of patches, as usual.
> > > >
> > > > Fwiw I can help as well -- starting next week. I can't do much this
week though.
> > > >
> > > > I would suggest starting with the cfbot to mark anything that isn't
> > > > applying cleanly and passing tests (and looking for more than design
> > > > feedback) as Waiting on Author and reminding the author that it's
> > > > commitfest time and a good time to bring the patch into a clean
state.
> > >
> > > Sounds like a plan; I'll make a start on that today/tomorrow as I have
> > > some time.
> >
> > If I'm not wrong, Jacob used the CF app to bulk-mail people about
> > patches not applying and similar things. That seemed to work well, and
> > doesn't require sending mails to dozens of threads.
>
> I don't see anything like that in the CF app (though I may be looking in
the
> wrong place). I also don't see how it would be possible to filter on
patches
> not applying in cbfot, as AFAICT the former is not aware of the
latter.Also, having gone through all the cfbot items with non-applying
patches (single
red "X"), sending a reminder without checking further doesn't seem the right
thing tod do - in two cases the patch was not applying because it had
already
been committed, and with another the consensus was to return it with
feedback.
With others, it's obvious the threads were recently active and I don't
think a
reminder is necessary right now.

Please do however let me know if I should be doing something differently.

Anyway changes since yesterday:

Needs review: 164 -> 156
Waiting on Author: 64 -> 68
Ready for Committer: 22 -> 21
Committed: 43 -> 47
Withdrawn: 9 -> 9
Rejected: 1 -> 1
Returned with Feedback: 4 -> 5

Following entries are reported with patch apply failure, but it's less clear
(to me) whether a simple request to update the patch is what is needed at
this point, because e.g. the thread is long and complex, or has been fairly
inactive for a while:

- "AcquireExecutorLocks() and run-time pruning"
https://commitfest.postgresql.org/40/3478/

- "pg_stat_activity: avoid showing state=active with wait_event=ClientRead"
https://commitfest.postgresql.org/40/3760/

- "logical decoding and replication of sequences, take 2"
https://commitfest.postgresql.org/40/3823/

- "Lazy JIT IR code generation to increase JIT speed with partitions"
https://commitfest.postgresql.org/40/3071/

- "Collation version and dependency helpers"
https://commitfest.postgresql.org/40/3977/

- "Time-delayed logical replication subscriber"
http://cfbot.cputube.org/patch_40_3581.log

- "Fix checkpointer sync request queue problems"
https://commitfest.postgresql.org/40/3583/

- "Nonreplayable XLog records by means of overflows and >MaxAllocSize
lengths"
https://commitfest.postgresql.org/40/3590/

- "Transparent column encryption"
https://commitfest.postgresql.org/40/3718/

The following have all received requests to update the patch(s), and
have been set to "Waiting on Author" where not already done:

- "XID formatting and SLRU refactorings (Independent part of: Add 64-bit
XIDs into PostgreSQL 15)"
https://commitfest.postgresql.org/40/3489/
(-> patches already updated; changed back to "WfC")

- "Add index scan progress to pg_stat_progress_vacuum"
https://commitfest.postgresql.org/40/3617/

- "Adding CommandID to heap xlog records"
https://commitfest.postgresql.org/40/3882/

- "Add semi-join pushdown to postgres_fdw"
https://commitfest.postgresql.org/40/3838/

- "Completed unaccent dictionary with many missing characters"
https://commitfest.postgresql.org/40/3631/

- "Data is copied twice when specifying both child and parent table in
publication"
https://commitfest.postgresql.org/40/3623/

- "Fix ExecRTCheckPerms() inefficiency with many prunable partitions "
https://commitfest.postgresql.org/40/3224/

- "In-place persistence change of a relation"
https://commitfest.postgresql.org/40/3461/

- "Move SLRU data into the regular buffer pool"
https://commitfest.postgresql.org/40/3514/

- "New [relation] options engine"
https://commitfest.postgresql.org/40/3536/

- "Nonreplayable XLog records by means of overflows and >MaxAllocSize
lengths"
https://commitfest.postgresql.org/40/3590/

- "Page compression for OLTP"
https://commitfest.postgresql.org/40/3783/

- "Provide the facility to set binary format output for specific OID's per
session"
https://commitfest.postgresql.org/40/3777/

- "Reducing power consumption when idle"
https://commitfest.postgresql.org/40/3566/

- "Reuse Workers and Replication Slots during Logical Replication"
https://commitfest.postgresql.org/40/3784/

- "Skip replicating the tables specified in except table option"
https://commitfest.postgresql.org/40/3646/

- "Teach pg_waldump to extract FPIs from the WAL stream"
https://commitfest.postgresql.org/40/3628/

- "[PATCH] Equivalence Class Filters"
https://commitfest.postgresql.org/40/3524/

- "improve handling for misconfigured archiving parameters"
https://commitfest.postgresql.org/40/3933/

- "logical decoding and replication of sequences, take 2"
https://commitfest.postgresql.org/40/3823/

Following open entries had actually already been committed:

- Improve description of XLOG_RUNNING_XACTS
https://commitfest.postgresql.org/40/3779/

- "Add native windows on arm64 support"
https://commitfest.postgresql.org/40/3561/

Following entry was set to "Returned with feedback":

- "On client login event trigger"
https://commitfest.postgresql.org/40/2900/

Regards

Ian Barwick