On Mon, Sep 22, 2025 at 2:15 PM Tom Lane wrote:
> I found time finally to look through all of these. Some notes:
Thanks. Committed 0001 and 0002 together. I agree with you that 0002
is a bit weird, but I think it is less weird than the status quo ante,
and it seems you agree. I think it's best t
I wrote:
> ... also that WinGetFuncArgInPartition is dereferencing
> a possibly-null "isout" pointer at several places
Looking around, there is only one in-core caller of
WinGetFuncArgInPartition, and it does pass a valid "isout" pointer,
explaining why this inconsistency wasn't obvious in testing
On Wed, 2025-09-24 at 11:52 -0500, Nathan Bossart wrote:
> On Wed, Sep 24, 2025 at 12:13:34PM -0400, Tom Lane wrote:
> > Nathan Bossart writes:
> > > * RangeVarCallbackForReindexIndex() was checking privileges on
> > > the table
> > > before locking it, so I reversed it in 0002.
> >
> > Don't we
Dear Amit, Ashutosh,
> Thanks for the analysis. I think we should go with the simple patch
> proposed in this thread to improve this test and that too only for
> HEAD.
To confirm, are you suggesting something like attached? It initially creates MVs
on both instances, then ensure REFRESH MATERIAL
Hi Wenhui,
Thanks for looking into this.
On Mon, Oct 13, 2025 at 5:41 PM wenhui qiu wrote:
>
> Hi Xuneng Zhou
>
>
> > - /* Unlock and release buffer */
> > - LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
> > - ReleaseBuffer(buffer);
> > + UnlockReleaseBuffer(buffer);
> > }
> Thanks for your patch! J
From the file header comment:
* rbtree.c
*implementation for PostgreSQL generic Red-Black binary tree package
*Adopted from http://algolist.manual.ru/ds/rbtree.php
*
* This code comes from Thomas Niemann's "Sorting and Searching Algorithms:
* a Cookbook".
*
* See http://www.cs.auck
Hi,
On Tue, 2025-10-07 at 10:44 +0200, Álvaro Herrera wrote:
> I wonder if anyone other than ancient forgotten RPM .spec files
> compiles these files anymore.
I feel like I should rewrite the main spec file from scratch. Will look.
> Anyway, I've fixed it now, thanks for reporting.
Thanks a lot
Hi,
On Mon, 13 Oct 2025 at 15:30, Dave Page wrote:
>
> I recently tried to update the version of Meson used in my winpgbuild
> project, which builds PostgreSQL for Windows, along with (almost) all the
> various dependencies. Something has changed in versions from 1.8.3 and above
> which causes
>> Attached is the v2 patch.
>>
>
> Thanks! This passes regression, and reads right to my eye and (most
> important to me) allows PostGIS to run under Pg19 again.
Thank you for the review! I have just pushed the v2 patch.
Best regards,
--
Tatsuo Ishii
SRA OSS K.K.
English: http://www.sraoss.co.j
Chao Li writes:
> After researching, I think we can take the similar way against constant and
> index for view. After altering a column’s type, related constrains and
> indexes will be rebuilt. Column type change may also break constraints or
> indexes. For example, if an int typed column has a
On Wed, Oct 1, 2025 at 1:09 PM Hayato Kuroda (Fujitsu)
wrote:
>
> Dear Peter,
>
> > IMO, it's not good to fool people into thinking something has happened
> > when in fact nothing happened at all. I think the logging of this tool
> > should be much more explicit wrt when it is/isn't in dry-run mod
Hello,
Barring further commentary, I intend to get 0001 committed tomorrow, and
0002 some time later -- perhaps by end of this week, or sometime next
week.
Regards
--
Álvaro HerreraBreisgau, Deutschland — https://www.EnterpriseDB.com/
Hi hackers,
I noticed that GUCs io_combine_limit and io_max_combine_limit are marked
as GUC_UNIT_BLOCKS, but in the documentation nothing is said about that.
Other GUCs marked as GUC_UNIT_BLOCKS have a phrase "If this value is
specified without units, it is taken as blocks, that is BLCKSZ bytes,
t
On Thu, Oct 9, 2025 at 4:14 PM Chao Li wrote:
>
> Hi Jian,
>
> Thanks for the patch. After reviewing it, I got a few small comments:
>
> On Oct 9, 2025, at 15:10, jian he wrote:
>
> Please check the attached v16.
>
> 1
> ```
> + List *partitions; /* oid list of partition oid for copy to */
> `
Hi,
just a second late :( i was about to post a patch addressing the refactors
which Robert mentioned ,anyway will have a look at your latest patch John
thanks :), curious about the tap test.
while i was writing the patch something suddenly struck me , that is why we
are even depending on last_c
On Wed, Oct 8, 2025 at 7:01 AM Álvaro Herrera wrote:
>
> On 2025-Oct-09, David Rowley wrote:
>
> > On Thu, 9 Oct 2025 at 00:36, Álvaro Herrera wrote:
> > > Yeah, that change makes sense to me. With an eye towards not forcing
> > > the translators to understand the message context or forced to tr
Re: Tomas Vondra
> It's probably more about the kernel version. What kernels are used by
> these systems?
It's the very same kernel, just different docker containers on the
same system. I did not investigate yet where the problem is coming
from, different libnuma versions seemed like the best bet.
Hi,
Thank you for your comments.
> On Mon, Oct 06, 2025 at 01:59:08AM +, Hayato Kuroda (Fujitsu) wrote:
> > Per my understanding, we already have a facility that terminates a
> background
> > worker, TerminateBackgroundWorker(). So, I'm afraid your proposal has
> already
> > been done by comb
On Tue, Oct 7, 2025 at 9:26 AM Nathan Bossart wrote:
>
> On Tue, Oct 07, 2025 at 11:50:46AM -0400, Robert Treat wrote:
> > On Tue, Oct 7, 2025 at 11:04 AM Nathan Bossart
> > wrote:
> >> I wonder if we should also add "aggressive".
> >
> > I don't think so. I feel like the point of the mode is to
On Tue, Sep 30, 2025 at 3:11 PM shveta malik wrote:
>
> On Tue, Sep 30, 2025 at 12:17 PM Fabrice Chapuis
> wrote:
> >
> > Hi,
> >
> > Here the generated v2 of the Patch.
> >
>
> Thanks. I have refactored the code for synchronize_one_slot() as there
> was some code-repetition. Please take it if yo
Hi,
On Mon, Oct 13, 2025 at 10:07 AM Xuneng Zhou wrote:
>
> Hi hackers,
>
> While reading the code related to streaming reads and their current
> use cases, I noticed that pgstatindex could potentially benefit from
> adopting the streaming read API. The required change is relatively
> simple—simi
On Thu, Sep 11, 2025 at 11:33 PM Chao Li wrote:
>
> I don’t understand why the two versions are different:
>
> if (XLogNeedsFlush(lsn))
> {
> /*
> * Remove the dirty buffer from the ring; necessary to prevent an
> * infinite loop if all ring members are dirty.
> */
> strategy->buffers[strategy->cu
On Sat, Oct 04, 2025 at 09:24:32PM +0530, Amit Kapila wrote:
> In the 0001 patch, pg_get_sequence_data() exposes two new fields
> log_cnt and page_lsn. I see that the later subscriber-side patch uses
> both, the first one in SetSequence(). It is not clear from the
> comments or the commit message o
On Wed, 15 Oct 2025 at 15:20, Peter Eisentraut wrote:
> On 06.10.25 20:05, Jacob Champion wrote:
> > I started on a proof of concept and very quickly hit a fork. Do I
> > 1) introduce a completely new config file, or
> > 2) adapt pg_service.conf to this use case?
>
> I've been thinking about this
> On Oct 17, 2025, at 00:27, Jacob Champion
> wrote:
>
> On Wed, Oct 15, 2025 at 7:30 PM Chao Li wrote:
>> The error message showed the correct line number, but “syntax error” is
>> confusing. Shouldn’t it be something like “invalid configuration parameter:
>> default”?
>
> I think so; se
On Thu, Oct 9, 2025 at 9:02 AM Robert Haas wrote:
> I understand that, but you mentioned a few different settings ...
> which could hypothetically turn into changing the default for 1 of
> them every year or two.
I think we should avoid equating "we can change it more frequently
than never" with
On Fri, Oct 10, 2025 at 8:09 AM Mario González Troncoso
wrote:
> Can you confirm you still have this problem on current master? I tried
Thanks for looking! Yes.
> to reproduce it in debian 12 by installing `postgresql-server-dev-15`
> and adding the -Dextra_include_dirs
The problem is with lib
Hi all,
While hacking on a different thing that touched pg_combinebackup, I
have bumped into a silly bug.
To keep it short, the version number is calculated based on this code
in read_pg_version_file(), where "version" is the result of strtoul()
applied to the contents of PG_VERSION:
return
On Thu, Oct 16, 2025 at 3:28 PM Renzo Dani wrote:
> I think I did it the wrong way with the patches.
> So here again with all in one file.
Thanks for updating the patch! I've pushed the patch.
Regards,
--
Fujii Masao
> On Oct 14, 2025, at 08:36, Tom Lane wrote:
>
> Chao Li writes:
>> I tested DEFAULT_IO_BUFFER_SIZE with 4K, 32K, 64K, 128K and 256K. Looks like
>> increasing the buffer size doesn’t improve the performance significantly.
>> Actually, with the buffer size 64K, 128K and 256K, the test results
On 10/8/25 02:04, Tomas Vondra wrote:
>
>
> On 10/7/25 23:08, Peter Geoghegan wrote:
>> On Tue, Oct 7, 2025 at 3:46 PM Andres Freund wrote:
>>> I think this discrepancy is largely due to the fact that Tomas' is testing
>>> with a cold cache (he has numbers for both), whereas most production
>>>
On Tue, Oct 07, 2025 at 05:36:11PM -0400, Greg Burd wrote:
> I put the patch aside for a while, then this past week at PGConf.dev/NYC
> I heard interest from a few people (Jeff Davis, Nathan Bossart) who
> encouraged me to move the code executing the expressions to just before
> acquiring the lock
Daniel Gustafsson writes:
>> On 13 Oct 2025, at 23:25, Chao Li wrote:
>> Fixed a typo: "iff" -> "if" that I found while reviewing the other patch.
> iff is shorthand for "if and only if", so it is correct here. That being
> said,
> it's also the cause of many typofix suggstions so it's clearly
On Fri, Sep 26, 2025 at 8:12 AM Robert Haas wrote:
> Basically anything that would allow us to do multiple tests
> without having to fork a new process for every single one seems like
> it would probably save quite a bit.
Agreed, it's pretty inefficient. (For comparison, it runs in about two
seco
On Fri, Sep 26, 2025 at 11:23 PM Tom Lane wrote:
> Richard Guo writes:
> > Looking at choose_plan_name(), IIUC, the nested loop is used to find
> > the next unused suffix number for a given name. I'm wondering why not
> > simply iterate through glob->subplanNames once, check the suffix
> > numbe
Hello Peter!
There is a code in vci_ros.c that initializes xl_heap_inplace xlrec.
Comment says this code was taken from src/backend/access/heap/heapam.c.
It was fine for Postgres 17 and earlier however struct xl_heap_inplace
has 6 fields, not one since commit 243e9b40f1b2. So nmsgs field of
xlrec
Hi Michael,
Thanks for the context. I missed the earlier discussions about the same
issue. Using unlink in the error paths makes sense to me. There is an edge
case in my mind, in case unlink fails as well, and we end up in the same
condition; however, the chance of that occurring is sufficiently l
Chao Li writes:
>> On Oct 14, 2025, at 08:36, Tom Lane wrote:
>> The thing we are really interested in here is how fast pg_restore
>> can skip over unwanted table data in a large archive file, and that
>> I believe should be pretty sensitive to block size.
> Not sure if I did something wrong, bu
On Wed, Oct 8, 2025 at 7:39 AM Robert Haas wrote:
>
Thanks for the feedback!
> I think the down side of this kind of system is that it makes life
> harder for people who want to write code that works everywhere. You
> have to take into account the possibility that the configuration file
> could
Hi,
On 2025-10-08 22:09:54 -0400, Andres Freund wrote:
> On October 8, 2025 10:05:59 PM EDT, Amit Langote
> wrote:
> >On Thu, Oct 9, 2025 at 2:15 AM Andres Freund wrote:
> >> On 2025-10-08 13:39:14 +1300, Thomas Munro wrote:
> >> > On Fri, Sep 19, 2025 at 11:44 AM Thomas Munro
> >> > wrote:
>
Hi,
On Sun, Oct 12, 2025 at 03:20:07PM +0200, Michael Banck wrote:
> I'll try to reproduce it here with the above --max-connections=10,
> thanks.
That didn't go so well. It ran for over 100 iterations of make check
without a fault (using EXTRA_REGRESS_OPTS="--max-connections=10"), so
not sure wha
Tom Lane 于2025年9月20日周六 00:16写道：
> Tender Wang writes:
> > While working on another issue, I stepped into
> match_orclause_to_indexcol().
> > I found below codes:
>
> > /* Only the operator returning a boolean suit the transformation. */
> > if (get_op_rettype(opno) != BOOLOID)
> > break;
>
On Thu, Oct 09, 2025 at 04:00:24AM +, Hayato Kuroda (Fujitsu) wrote:
> Based on that, I imagined that we could launch a bgworker and create
> another database by using template. Or is it already handled?
Oh, I was not following you here. Yes, we should have something with
a bgworker connected
For failover to work correctly, if someone changes the GUC
synchronous_standby_names to enable sync replication, then we need to
understand the exact moment when backends will begin to block in order
to correctly determine when we can failover without data loss.
There's an older mailing list threa
On 10/6/25 17:14, Andres Freund wrote:
> Hi,
>
> On 2025-10-06 02:59:16 +0200, Tomas Vondra wrote:
>> So I decided to try doing this on a couple different devices, and see
>> what random_page_cost values that gives me. Attached is a script doing
>> such benchmark:
>>
>> (1) initializes a new cl
On 2025-Oct-06, Nazir Bilal Yavuz wrote:
> Sorry for the confusion. We have a script that installs linux
> dependencies for the CI [1]. I just copied the contents of this script
> to make sure that I don't miss any dependencies, probably these
> dependencies are much larger than we actually need.
On Thu, 9 Oct 2025 at 00:16, Masahiko Sawada wrote:
>
> On Wed, Oct 8, 2025 at 11:39 AM vignesh C wrote:
> >
> > On Tue, 7 Oct 2025 at 23:40, Masahiko Sawada wrote:
> > >
> > > On Mon, Oct 6, 2025 at 6:55 PM Euler Taveira wrote:
> > > >
> > > > On Mon, Oct 6, 2025, at 5:00 PM, Masahiko Sawada w
[ sorry for ridiculously slow response to this ]
Tomas Vondra writes:
> Here's a patch trying to do it more like this - by manipulating the
> lists describing the join problems, before it's passed the the actual
> join search algorithm (which is where the PoC patch did this).
> I wonder if this i
Thank you Arseniy for the thoughtful and detailed feedback, I have
addressed the concerns in the new patch.
On Wed, Oct 8, 2025 at 3:07 AM Arseniy Mukhin
wrote:
> Yes, we hold the lock while we are writing the commit record to wal,
> but other transactions don't see our transaction as committed
> On Oct 8, 2025, at 22:53, Joel Jacobson wrote:
>
>> 1
>> ```
>> + channels = GetPendingNotifyChannels();
>> +
>> LWLockAcquire(NotifyQueueLock, LW_EXCLUSIVE);
>> - for (ProcNumber i = QUEUE_FIRST_LISTENER; i != INVALID_PROC_NUMBER; i
>> = QUEUE_NEXT_LISTENER(i))
>> + foreach(lc, channels)
>>
Hi,
On Fri, Sep 26, 2025 at 6:19 PM Jim Jones wrote:
>
> Given that the lack of tests allowed this bug to go undetected until
> now, I'd suggest to include additional tests in this patch to prevent
> similar issues in the future. Something like 0002 attached. What do you
> think?
>
Thanks for th
> On Oct 13, 2025, at 15:18, Daniel Gustafsson wrote:
>
>> On 13 Oct 2025, at 07:34, Chao Li wrote:
>>
>> Hi Hacker,
>>
>> I just noticed $SUBJECT:
>>
>> In doc/src/sgml/postgres.sgml, line 12:
>> ```
>> https://postgr.es/c/";>
>> ```
>>
>> https://postgr.es/c/ is resolved to
>> https://g
Am 30.09.25 um 11:36 schrieb Michael Banck:
Hi,
On Tue, Sep 30, 2025 at 11:17:07AM +0200, Stefanie Janine Stölting wrote:
Am 30.09.25 um 10:50 schrieb Thom Brown:
On Tue, 30 Sept 2025 at 09:12, Stefanie Janine Stölting
wrote:
Hi there,
I installed a fresh Debian 13 and added the packages
f
On Tue, Oct 07, 2025 at 10:17:38AM -0700, Masahiko Sawada wrote:
> Right. I think we cannot display both things in one mode column. Since
> both manual vacuums and anti-wraparound autovacuums can enter the
> failsafe mode dynamically, if we show "failsafe" in the mode column,
> we would lose the in
On Mon, Oct 6, 2025 at 10:22 AM shveta malik wrote:
>
> On Thu, Oct 2, 2025 at 4:53 PM Ajin Cherian wrote:
> >
> >
> > Attaching patch v15 addressing these comments.
> >
>
> It seems v15 had a compilation issue. Resolved it. Attaching v15 again.
>
Verified patch, works well.
I have changed doc
> On 6 Oct 2025, at 04:42, Jeremy Schneider wrote:
>
> A large and stressed system could get into a state where fsync takes
> awhile.
For such cases I would like to have a function that ensures walreceiver will
not acknowledge any new LSN. And then returns last ack'ed LSN.
Today we have to
On Sun, 5 Oct 2025 at 13:56, David Rowley wrote:
> Another alternative that I'm thinking about which might be better is
> to double-down on the varno==0 and invent a special varno and define
> SETOP_VAR. I'd feel better about doing that as I didn't feel good
> about coding the magic number for the
On 27/5/2024 20:20, Michael Paquier wrote:
Please note that I've been studying ways to have pg_stat_statements
being plugged in directly with the shared pgstat APIs to get it backed
by a dshash to give more flexibility and scaling, giving a way for
extensions to register their own stats kind. In
Hi Jian,
On Thu, Oct 9, 2025 at 12:48 PM jian he wrote:
> On Wed, Oct 8, 2025 at 12:10 PM Amit Langote wrote:
> >
> > After sleeping on this, I realized the right fix is not to strip
> > CollateExpr in transformJsonBehavior(), but to stop recursing on
> > JsonBehavior.expr in exprSetCollation().
The following review has been posted through the commitfest application:
make installcheck-world: tested, passed
Implements feature: tested, passed
Spec compliant: tested, passed
Documentation:not tested
Greetings,
I took a look at the v2 patch for the PGXS update-po
On Sun, 24 Aug 2025 at 22:12, Andrew Dunstan wrote:
>
>
> On 2025-08-23 Sa 9:08 PM, Noah Misch wrote:
>
> On Wed, Jul 30, 2025 at 02:51:59PM -0400, Andrew Dunstan wrote:
>
> OK, now that's reverted we should discuss how to proceed. I had two thoughts
> - we could use invent a JSON format for the g
Hi Álvaro,
Thanks for your feedback.
> Hmm, but instead of keeping the comment about why this is bogus, why not
> just fix it and remove the comment? You could do something like
>
> nsp = get_namespace_name( .. );
> diffname = psprintf("%s_%s_%d", nsp, RelationGetRelationName( .. ), 2);
> tempna
On Mon, Oct 13, 2025 at 11:43 PM Michael Paquier wrote:
>
> On Tue, Oct 14, 2025 at 08:31:04AM +0500, Kirill Reshke wrote:
> > Hi! Should we also bump XLOG_PAGE_MAGIC after d96f87332 & add323da40a
> > or do we wait for full set to be committed?
>
> I may be missing something, of course, but d96f87
On Wed, Oct 15, 2025 at 7:30 PM Chao Li wrote:
> The error message showed the correct line number, but “syntax error” is
> confusing. Shouldn’t it be something like “invalid configuration parameter:
> default”?
I think so; see patch 0003 in [1].
--Jacob
[1]
https://postgr.es/m/CAOYmi%2BkYKPX
Hi all,
I've noticed that many TAP tests in the codebase make sub-optimal use of
the "ok()" function. Specifically, ok() is often used for expressions
involving comparison operators or regex matches, which is not ideal because
other Test::More functions provide much clearer diagnostic messages whe
Chao Li writes:
> I tested DEFAULT_IO_BUFFER_SIZE with 4K, 32K, 64K, 128K and 256K. Looks like
> increasing the buffer size doesn’t improve the performance significantly.
> Actually, with the buffer size 64K, 128K and 256K, the test results are very
> close. I tested both with lz4 and none comp
Hi,
On 2025-09-30 10:42:00 +0200, Philipp Marek wrote:
> Here's the patch again, this time with a 128kB buffer size.
>
> This gives us nearly the same gains (~7%) for the blob INSERTs,
> and the additional memory usage (120kB) shouldn't really matter,
> with "temp_buffer"s 8MB and "work_mem" 4MB
> On 25 Sep 2025, at 09:23, Hayato Kuroda (Fujitsu)
> wrote:
>
> Dear hackers,
>
>> While reviewing others, I found $SUBJECT.
>
> I found another cleanup point related with this. In CreateCheckPoint():
>
> ```
> #ifdef USE_INJECTION_POINTS
> INJECTION_POINT("checkpoint-before-old-wal-removal"
On Thu, Sep 25, 2025 at 4:25 AM Amul Sul wrote:
> > Another thing that isn't so nice right now is that
> > verify_tar_archive() has to open and close the archive only for
> > init_tar_archive_reader() to be called to reopen it again just moments
> > later. It would be nicer to open the file just o
I wrote:
> David Rowley writes:
>> I'd be happy for someone else to
>> continue. I'm a bit hesitant due to the "#if defined(NAN) &&
>> !(defined(__NetBSD__) && defined(__mips__))", which are systems I have
>> little knowledge of.
> Fair enough, I'll take responsibility for that bit. Might as wel
On Tue, Oct 14, 2025 at 08:31:04AM +0500, Kirill Reshke wrote:
> Hi! Should we also bump XLOG_PAGE_MAGIC after d96f87332 & add323da40a
> or do we wait for full set to be committed?
I may be missing something, of course, but d96f87332 has not changed
the WAL format, VISIBILITYMAP_ALL_VISIBLE and VI
On Sat, Sep 20, 2025 at 4:59 AM Amit Kapila wrote:
>
> On Thu, Sep 18, 2025 at 11:46 PM Masahiko Sawada
> wrote:
> >
> > On Thu, Sep 18, 2025 at 1:33 AM Amit Kapila wrote:
> > >
> > > If we compare conflict_history_table with the slot that gets created
> > > with subscription, one can say the s
On Tue, Sep 23, 2025 at 7:01 AM Christoph Berg wrote:
> Libs handling is ok:
>
> $ pkgconf --libs libpq
> -lpq
>
> But CFLAGS is not:
>
> $ pkgconf --cflags libpq
> Package libcurl was not found in the pkg-config search path.
> Perhaps you should add the directory containing `libcurl.pc'
> to the
Thank you for your path ，The path LGTM
On Wed, 1 Oct 2025 at 16:15, Zhang Mingli wrote:
> Hi,
>
> The comment incorrectly cited the defunct function BufFileOpenShared()
> which is replaced in commit dcac5e7ac157964f71f15d81c7429130c69c3f9b.
>
> Update the reference to the current function: BufFi
Hi,
On 2025-10-09 11:43:57 -0400, Robert Haas wrote:
> While $SUBJECT today -- using the sources from
> https://git.postgresql.org/git/pg_bsd_indent.git -- I ran into some
> problems. I got errors due to the replacement of
> pg_attribute_noreturn() with pg_return(), and warnings due to variable
>
On 2025-Oct-09, Antonin Houska wrote:
> + /*
> + * Make sure ANALYZE is specified if a column list is present.
> + */
> + if ((params->options & CLUOPT_ANALYZE) == 0 && stmt->relation->va_cols
> != NIL)
> + ereport(ERROR,
> + (errcode(ERRC
hi. I still can not compile your v2.
../../Desktop/pg_src/src1/postgres/src/backend/utils/adt/ruleutils.c:
In function ‘get_formatted_string’:
../../Desktop/pg_src/src1/postgres/src/backend/utils/adt/ruleutils.c:13770:9:
error: function ‘get_formatted_string’ might be a candidate for
‘gnu_printf’
On Fri, Oct 10, 2025 at 6:24 AM Robert Haas wrote:
> That's a really interesting observation. I've always found it a bit
> odd that we put things like sslca and sslrootcert into the connection
> string, so I think you have a point, here. Not sure I agree about
> sslnegotiation or gssencmode, thoug
On 10/16/25 13:38, Christoph Berg wrote:
>> src/test/regress/expected/numa.out | 13 +++
>> src/test/regress/expected/numa_1.out | 5 +
>
> numa_1.out is catching this error:
>
> ERROR: libnuma initialization failed or NUMA is not supported on this
> platform
>
> This is what I'm
On Tue, Oct 14, 2025 at 01:13:38PM +0200, Matěj Klonfar wrote:
> I can imagine this limitation is likely a holdover from the system's
> evolution from physical replication where comments make no sense. However,
> in logical replication walsender mode both SQL and replication statements
> can be iss
Re: To Tomas Vondra
> So maybe all that's needed is a get_mempolicy() call in
> pg_numa_available() ?
Or perhaps give up on pg_numa_available, and just have two _1.out and
_2.out that just contain the two different error messages, without
trying to catch the problem.
Christoph
On Tue, Oct 7, 2025 at 10:53 AM shveta malik wrote:
>
> 2)
> postgres=# create publication pub1 for all sequences WITH(publish='insert');
> ERROR: publication parameters are not supported for publications
> defined as FOR ALL SEQUENCES
>
> postgres=# alter publication pub1 add table tab1;
> ERROR
Hi all,
I have spotted a couple of buildfarm failures for buildfarm member
phycodorus on REL_14_STABLE and REL_13_STABLE:
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=phycodurus&dt=2025-10-15%2009%3A12%3A36
These are sporadic, pointing at a backtrace with JIT in some cases on
REL_13_ST
On Tue, Oct 14, 2025 at 10:01:37AM -0700, Jeff Davis wrote:
> If we start with an OID, what's the right way to do these kinds of
> checks? Could we do an ACL check, then lock it, then do an ACL check
> again to catch OID wraparound?
I tried something like this upthread [0]. My feeling was that th
Hi Nathan,
> Seems reasonable to me. Most of these seem to be recent additions since
> last year's round of IWYU commits. Any reason to limit this to
> src/backend/commands? Why not tackle the whole tree at once?
The main problem here is that clangd is a language server, which means
I see erro
Hi Akshay,
When applying the patch, I got a number of errors and the tests failed. I
think it stems from:
+ targetTable = relation_open(tableID, NoLock);
+ relation_close(targetTable, NoLock);
I changed them to use "AccessShareLock" and it worked, and it's probably
relevant to change table_clos
Hi,
The comment incorrectly cited the defunct function BufFileOpenShared()
which is replaced in commit dcac5e7ac157964f71f15d81c7429130c69c3f9b.
Update the reference to the current function: BufFileOpenFileSet().
--
Zhang Mingli
HashData
v0-0001-Fix-incorrect-function-reference-in-comment.pat
On Mon, Oct 6, 2025 at 11:24 AM Tomas Vondra wrote:
>
> On 10/6/25 07:29, Tom Lane wrote:
> > Tomas Vondra writes:
[..]
>
> > Another angle is that I expect that the ongoing AIO work will largely
> > destroy the existing model altogether, at least if you think in terms
> > of the model as trying
Jim Jones writes:
> [ v3-0001-Disallow-ATOMIC-functions-depending-on-temp-relat.patch ]
Got around to reading the patch finally. I don't like anything
about this implementation. It introduces yet another place that
(thinks it) knows how to find all the dependencies in a query
tree, requiring ye
Hi,
7% is an interesting improvement for these cases. I see it as a nice
feature,
What about adding a GUC variable to have the ability to configure
PQ_RECV_BUFFER_SIZE based on the user's needs?
-Filip-
út 30. 9. 2025 v 10:42 odesílatel Philipp Marek
napsal:
> Here's the patch again, this
On Tue, Oct 14, 2025 at 07:21:13PM +0900, Tatsuo Ishii wrote:
> V2 patch pushed. Thanks.
Coverity thinks that this code has still some incorrect bits, and I
think that it is right to think so even on today's HEAD at
02c171f63fca.
In WinGetFuncArgInPartition()@nodeWindowAgg.c, we have the followin
On Wed, Oct 08, 2025 at 05:30:58PM +0300, Aleksander Alekseev wrote:
>> clangd indicates that certain #include's are redundant. Removing them
>> will speed up the build process a bit.
>
> OK, cfbot tells me that touching collationcmds.c was a mistake. Here
> is a corrected patch.
Seems reasonable
On Thu, Oct 9, 2025 at 4:38 PM Chao Li wrote:
>
> I think is patch is helpful. A few comments:
>
> On Oct 9, 2025, at 08:55, Peter Smith wrote:
>
>
>
>
>
> 1 - 0001
> ```
> + if (dryrun)
> + {
> + pg_log_info("-");
> + pg_log_info("pg_archivecl
On Fri, Oct 10, 2025 at 3:44 PM Nathan Bossart wrote:
> On Fri, Oct 10, 2025 at 02:42:57PM -0400, Robert Haas wrote:
> > I think this is a reasonable starting point, although I'm surprised
> > that you chose to combine the sub-scores using + rather than Max.
>
> My thinking was that we should cons
Hi Jim,
One last thing: should we perhaps check for NULL before calling
> appendStringInfo here?
>
> /* pg_get_triggerdef_worker retrieves the trigger definition */
> res = pg_get_triggerdef_worker(trgOid, false);
>
> appendStringInfo(&buf, "%s;", res);
>
Yes, you're correct. I've added that in
On Tue, Sep 23, 2025 at 7:18 AM Nathan Bossart wrote:
> I would have used this in the DSM registry if it was available. Patch
> attached.
Likewise for libpq-oauth, as attached.
Thanks,
--Jacob
0001-Make-some-use-of-anonymous-unions-libpq-oauth.patch
Description: Binary data
On 2025-Sep-28, Tom Lane wrote:
> =?UTF-8?Q?=C3=81lvaro_Herrera?= writes:
> > The changes to inet.h and varbit.h are surprising, but required.
>
> Why? Those headers compile now, and you've not removed any includes
> that they might indirectly depend on.
That's true.
> If what you mean is tha
On Fri, 2025-10-10 at 11:26 -0500, Nathan Bossart wrote:
> On Thu, Oct 09, 2025 at 04:18:03PM -0500, Nathan Bossart wrote:
> > There's a similar pattern in get_rel_from_relname() in dblink.c,
> > which also
> > seems to only be used with an AccessShareLock (like pg_prewarm).
> > My best
> > guess
David Rowley writes:
> As for fixing it; testing for a Const-false havingClause can't be done
> as that only works for this case which const-folding happens to figure
> out during planning. You could equally have something Var-less like:
Yeah, I don't think the havingClause being constant-false i
I was just looking at the VACUUM option processing and I saw that the
code to process the PARALLEL option doesn't make use of the code in
defGetInt32() that's meant to handle empty and non-integer parameters.
ExecVacuum() has code to handle an empty PARALLEL parameters, but not
non-integer ones. Th
1 - 100 of 530 matches
Mail list logo
