tries in ~53 buckets, but with 1.0 you expect
to have 100 entries in ~64 buckets. It'd be a fair criticism that
that's arbitrarily different to the choice made for hash joins, and
dynahash's default is 1 (though it's a run-time parameter).
[1]
https://math.stackexchange.com/questions/470662/average-number-of-bins-occupied-when-throwing-n-balls-into-n-bins
--
Thomas Munro
http://www.enterprisedb.com
On Sat, Oct 27, 2018 at 10:03 AM Thomas Munro
wrote:
> space-vs-collision trade-off. If I have my maths right[1], with 0.75
> you expect to have 75 entries in ~53 buckets, but with 1.0 you expect
> to have 100 entries in ~64 buckets. It'd be a fair criticism that
Forgot to add
On Sun, Oct 28, 2018 at 1:22 AM Tomas Vondra
wrote:
> On 10/27/2018 01:51 PM, Antonin Houska wrote:
> > Antonin Houska wrote:
> >> Thomas Munro wrote:
> >>> On Sat, Oct 27, 2018 at 12:30 AM Antonin Houska wrote:
> >>> Are you saying there is a b
>> I have a access to too old 11.2 Oracle. There I had to modify query
>> because there is not boolean type. I replaced bool by int, but I got a
>> error
>> ORA-19224:XPTY-004 .. expected node()*, got xs:string - it doesn't work
>> with/without string() wrappings.
>>
>The problem is in last line -
)
> It is currently in the list of patches for the current commitfest, but
> based on your previous post I'm not sure if you're planning to get back to
> this patch just now. If you plan to resume work on it, I'll sign up as a
> reviewer.
I'm also signed up to review.
--
Thomas Munro
http://www.enterprisedb.com
rms that has pread/pwrite.
Here's a patch to add Windows support by supplying
src/backend/port/win32/pread.c. Thoughts?
--
Thomas Munro
http://www.enterprisedb.com
0001-Use-pread-pwrite-instead-of-lseek-read-write-v8.patch
Description: Binary data
0002-Supply-pread-pwrite-implem
On Sun, Nov 4, 2018 at 12:03 AM Thomas Munro
wrote:
> On Sat, Nov 3, 2018 at 2:07 AM Jesper Pedersen
> wrote:
> > This still applies, and passes make check-world.
> >
> > I wonder what the commit policy is on this, if the Windows part isn't
> > included. I re
t; On 2018-Nov-04, Thomas Munro wrote:
> >> Here's a patch to add Windows support by supplying
> >> src/backend/port/win32/pread.c. Thoughts?
>
> > Hmm, so how easy is to detect that somebody runs read/write on fds where
> > pread/pwrite have occurred? I g
x27;m not very
> optimistic about this any more.
>
> But ripping it out in the back branches seems unnecessary. I'd just
> leave the bugs unfixed there. Most likely nobody is using that stuff,
> but if they are, let's not break it.
Thanks. Pushed to master only.
--
Thomas Munro
http://www.enterprisedb.com
On Tue, Nov 6, 2018 at 6:15 PM Michael Paquier wrote:
> On Tue, Nov 06, 2018 at 05:29:36PM +1300, Thomas Munro wrote:
> > Thanks. Pushed to master only.
>
> Just a wild idea while this thread is hot: could we add in the
> description of the broken APIs or in their header
in the docs part from v1.
I think PGSOCKETDIR should be mentioned in the documentation like the
other environment variables, and also I'm wondering if it actually
works: you set it to the current working directory first, then parse
the command line option if present, and then read the env var
other from COPY ... FROM PROGRAM. The differences were:
PIPE, TTIN, TTOU and USR2. For the first and last of those, the
default action would be to terminate the process, but the COPY PROGRAM
child ignored them; for TTIN and TTOU, the default action would be to
stop the process, but again they are ignored. Why do bgwriter.c,
startup.c, ... set SIGTTIN and SIGTTOU back to SIG_DFL, but not
regular backends?
--
Thomas Munro
http://www.enterprisedb.com
On Wed, Nov 7, 2018 at 4:42 AM Jesper Pedersen
wrote:
> Passes check-world, and includes the feedback on this thread.
>
> New status: Ready for Committer
Thanks! Pushed. I'll keep an eye on the build farm to see if
anything breaks on Cygwin or some other frankenOS.
--
Tho
better idea for how to do this. It occurred to me that I might
be missing something or this would have been done already...
--
Thomas Munro
http://www.enterprisedb.com
On Tue, Sep 25, 2018 at 2:09 PM Thomas Munro
wrote:
> Some people like to use DNS SRV records to advertise LDAP servers on
> their network. Microsoft Active Directory is usually (always?) set up
> that way. Here is a patch to allow our LDAP auth module to support
> that kind of di
f that address space as necessary (instead of doing that in
dsa_get_address()). AFAIK that would work, but it seems to be a bit
weird to go to such lengths. It would be a kind of home-made
simulation of threads. On the other hand, that is what we're already
doing in dsa.c, except more slowly due to extra software address
translation from funky pseudo-addresses.
3. Something something threads.
--
Thomas Munro
http://www.enterprisedb.com
there's no objection I will commit this.
Thanks for adjusting that. I suppose I would have known about this if
cfbot checked every patch with valgrind, which I might look into.
I'm a little confused about how an uninitialised value originating in
an OID list finishes up in an xlog buffer, considering that OIDs don't
have padding.
--
Thomas Munro
http://www.enterprisedb.com
that to be expected.
I tried to think of a scenario where the already-received output is
truly invalidated by a later error that happens after we close the
pipe. It could be something involving a program that uses something
optimistic like serializable snapshot isolation that can later de
On Fri, Oct 19, 2018 at 6:42 PM Craig Ringer wrote:
> On Fri, 19 Oct 2018 at 07:27, Thomas Munro
> wrote:
>> 2. I am +1 on back-patching Craig's PANIC-on-failure logic. Doing
>> nothing is not an option I like. I have some feedback and changes to
>> propose tho
On Tue, Nov 6, 2018 at 12:42 AM Kuntal Ghosh wrote:
> On Thu, Nov 1, 2018 at 8:53 AM Thomas Munro
> wrote:
> > It passes make check on Unix and Windows, though currently it's
> > failing some of the TAP tests for reasons I'm looking into (possibly
> > due to bu
On Fri, Nov 9, 2018 at 7:07 AM Robert Haas wrote:
> On Wed, Nov 7, 2018 at 9:41 PM Thomas Munro
> wrote:
> > My plan is do a round of testing and review of this stuff next week
> > once the dust is settled on the current minor releases (including
> > fixing a few typos
011%3A58%3A07
--
Thomas Munro
http://www.enterprisedb.com
On Fri, Nov 9, 2018 at 2:19 PM Ideriha, Takeshi
wrote:
> From: Thomas Munro [mailto:thomas.mu...@enterprisedb.com]
> >I know of 3 ideas that would make your idea C work, so that you could share
> >something as complicated as a query plan directly without having to
> >deseri
is not enough... due to buffering, it works. Try 100:
postgres=# create foreign table ft5 (a text) server svf1 options
(program 'ruby -e "for i in 1..100 do puts i; end"');
CREATE FOREIGN TABLE
postgres=# select * from ft5 limit 5;
ERROR: program "ruby -e "for i in 1..100 do puts i; end"" failed
DETAIL: child process exited with exit code 1
--
Thomas Munro
http://www.enterprisedb.com
k
> explicitly using barriers is much saner these days.
>
> Comments?
+1
I said the same over here:
https://www.postgresql.org/message-id/flat/CAEepm%3D1nff0x%3D7i3YQO16jLA2qw-F9O39YmUew4oq-xcBQBs0g%40mail.gmail.com
--
Thomas Munro
http://www.enterprisedb.com
On Mon, Nov 12, 2018 at 9:34 PM Noah Misch wrote:
> On Sat, Oct 13, 2018 at 11:45:17PM +1300, Thomas Munro wrote:
> > + /* Set a different seed for random() in every backend. */
> > + srandom((unsigned int) MyProcPid ^ (unsigned int) MyStartTimestamp);
>
> > -
On Fri, Jun 16, 2017 at 9:42 AM Thomas Munro
wrote:
> On Fri, Sep 25, 2015 at 5:06 PM, Pavel Stehule
> wrote:
> > 2015-09-25 0:25 GMT+02:00 Jim Nasby :
> >> On 9/24/15 3:35 PM, Peter Geoghegan wrote:
> >>>
> >>> I would worry about the
On Tue, Nov 13, 2018 at 9:45 AM Robert Haas wrote:
> On Thu, Nov 8, 2018 at 9:05 PM Thomas Munro
> wrote:
> > * I had some ideas about some kind of "allocation rollback" interface:
> > you begin an "allocation transaction", allocate a bunch of stuff
> &
t's a false positive.
> Is that the case, or should MAX_HBA_OPTIONS be increased?
Right, thank you. It seems clear that MAX_HBA_OPTIONS should be
increased and the comment near its definition is wrong. Will fix.
--
Thomas Munro
http://www.enterprisedb.com
On Tue, Nov 13, 2018 at 3:42 PM Thomas Munro
wrote:
> On Tue, Nov 13, 2018 at 3:02 PM Julian Hsiao wrote:
> > During a routine Coverity scan of our internal PostgreSQL fork, it
> > issued a buffer overrun warning for src/backend/libpq/hba.c,
> > gethba_options()[0
(Replies to a couple of different messages below)
On Wed, Nov 14, 2018 at 6:04 AM Robert Haas wrote:
> On Sun, Nov 11, 2018 at 9:59 PM Thomas Munro
> wrote:
> > There is one major problem with this patch
>
> If there's only one, you're doing great! Although admitt
s to be a major subproblem to deal with early in the project. I
will be very happy if we can come up with something :-)
On Wed, Nov 14, 2018 at 12:40 AM Ideriha, Takeshi
wrote:
> From: Thomas Munro [mailto:thomas.mu...@enterprisedb.com]
> >postgres=# call hoge_list(3);
> >NOTICE
On Wed, Nov 14, 2018 at 3:24 PM Noah Misch wrote:
> On Mon, Nov 12, 2018 at 09:39:01AM -0500, Tom Lane wrote:
> > Thomas Munro writes:
> > > On Mon, Nov 12, 2018 at 9:34 PM Noah Misch wrote:
> > >> Compared to the old code, the new code requires more wa
On Wed, Nov 14, 2018 at 6:34 PM Noah Misch wrote:
> On Wed, Nov 14, 2018 at 05:50:26PM +1300, Thomas Munro wrote:
> > On Wed, Nov 14, 2018 at 3:24 PM Noah Misch wrote:
> > > What counts is the ease of predicting a complete seed. HEAD's algorithm
> > > has
>
On Wed, Nov 14, 2018 at 8:52 PM Noah Misch wrote:
> On Wed, Nov 14, 2018 at 08:22:42PM +1300, Thomas Munro wrote:
> > On Wed, Nov 14, 2018 at 6:34 PM Noah Misch wrote:
> > > On Wed, Nov 14, 2018 at 05:50:26PM +1300, Thomas Munro wrote:
> > > > On Wed, Nov 14, 201
On Thu, Nov 15, 2018 at 4:25 AM Tom Lane wrote:
> Thomas Munro writes:
> > What do you think about the attached?
>
> I think you need to cast MyStartTimestamp to uint64 before shifting
> to ensure portable behavior of the shifts. In principle it wouldn't
> matter bec
On Fri, Nov 9, 2018 at 4:42 PM David Rowley
wrote:
> On 7 November 2018 at 11:46, Andres Freund wrote:
> > On 2018-11-07 11:40:22 +1300, Thomas Munro wrote:
> >> PostgreSQL likes to probe the size of relations with lseek(SEEK_END) a
> >> lot. For example, a
On Wed, Nov 7, 2018 at 4:39 PM Thomas Munro
wrote:
> On Tue, Sep 25, 2018 at 2:09 PM Thomas Munro
> wrote:
> > Some people like to use DNS SRV records to advertise LDAP servers on
> > their network. Microsoft Active Directory is usually (always?) set up
> > that way.
nt.event == WL_TIMEOUT. Oops.
I will fix that when I post a new experimental version that uses the
bgworker as discussed, and we can try to figure out if that design
will fly.
--
Thomas Munro
http://www.enterprisedb.com
to put it another way, you can't be given a lower sequence number
than another process that has already written, because that other
process must have been given a sequence number before it wrote.
--
Thomas Munro
http://www.enterprisedb.com
On Sat, Nov 17, 2018 at 5:56 AM Tom Lane wrote:
> Thomas Munro writes:
> > [ 0001-Add-WL_EXIT_ON_PM_DEATH-pseudo-event-v4.patch ]
>
> I took a quick look through this. I have no objection to the idea of
> letting the latch infrastructure do the proc_exit(1), but I'm won
On Fri, Nov 9, 2018 at 9:06 AM Robert Haas wrote:
> On Thu, Nov 8, 2018 at 3:04 PM Thomas Munro
> wrote:
> > My reasoning for choosing bms_join() is that it cannot fail, assuming
> > the heap is not corrupted. It simply ORs the two bit-strings into
> > whichever is the
On Fri, Nov 9, 2018 at 9:03 AM Thomas Munro
wrote:
> On Fri, Nov 9, 2018 at 7:07 AM Robert Haas wrote:
> > On Wed, Nov 7, 2018 at 9:41 PM Thomas Munro
> > wrote:
> > > My plan is do a round of testing and review of this stuff next week
> > > once the dus
show them as
different types and show the page size, but you'd have to check that
while it's actually running a parallel query. For example, a large
parallel hash join that runs for a while would do it, and in theory
you might be able to see a small performance improvement for larger
pag
an't use large pages on
Linux (because Linux's POSIX shm_open() really just opens files on
/dev/shm, which has a 4k block size), but can use them on FreeBSD
(because its shm_open() isn't bound to page sizes, it can and
sometimes decides to use large pages).
--
Thomas Munro
http://www.enterprisedb.com
On Wed, Nov 21, 2018 at 9:07 AM Andres Freund wrote:
> On 2018-11-21 09:00:58 +1300, Thomas Munro wrote:
> > On Wed, Nov 21, 2018 at 4:37 AM REIX, Tony wrote:
> > > YES ! Reading this file, your suggestion should work ! Thx !
> > >
> > > I've rebuilt an
On Wed, Nov 21, 2018 at 1:08 PM Andres Freund wrote:
> Let's see what I broke :/
Per rhinoceros, contrib/sepgsql. This patch fixes the build for me
(on Debian with libselinux1-dev installed), though I don't know how to
test that it works and make check does nothing useful in there
ostgresql.org/message-id/flat/CAPpHfds9trA6ipezK3BsuuOSQwEmESiqj8pkOxACFJpoLpcoNw%40mail.gmail.com#9b04576b717175e9dbf03cc991977d3f
--
Thomas Munro
http://www.enterprisedb.com
On Wed, Nov 21, 2018 at 5:43 PM Thomas Munro
wrote:
> Googling led me to some recently posted clues on SO and elsewhere
> about a flag FILE_RENAME_FLAG_POSIX_SEMANTICS which can be used with
> SetFileInformationByHandle(), that wasn't mentioned in the earlier
> -hackers thread
now if it's really safe. If that is somehow magically OK,
you'd probably also need the chdir() hack from proc_exit() to get
per-pid files.
--
Thomas Munro
http://www.enterprisedb.com
ure what is the suitable name.
I think that might be interesting information in the future, but let's
consider that for a later patch.
--
Thomas Munro
http://www.enterprisedb.com
0001-Don-t-count-zero-filled-buffers-as-read-in-EXPLAI-v2.patch
Description: Binary data
On Sat, Nov 17, 2018 at 2:27 PM Thomas Munro
wrote:
> Thanks for the review.
Pushed.
--
Thomas Munro
http://www.enterprisedb.com
ent variables or loader controls, since
apparently other things like data segments etc apparently can, though
I can't find any text that says that's the case and I have no AIX
system).
--
Thomas Munro
http://www.enterprisedb.com
On Sun, Nov 25, 2018 at 3:38 AM Christoph Berg wrote:
> Re: Thomas Munro 2018-11-23
> > Add WL_EXIT_ON_PM_DEATH pseudo-event.
>
> I think this broke something:
>
> TRAP: FailedAssertion(»!(!IsUnderPostmaster || (wakeEvents & (1 << 5)) ||
> (wakeEvents & (1
On Sun, Nov 25, 2018 at 12:59 PM Thomas Munro
wrote:
> On Sun, Nov 25, 2018 at 3:38 AM Christoph Berg wrote:
> > TRAP: FailedAssertion(»!(!IsUnderPostmaster || (wakeEvents & (1 << 5)) ||
> > (wakeEvents & (1 << 4)))«, Datei:
> > »/build/pos
ticed on cfbot.cputube.org:
pg_type.c:167:10: error: passing argument 3 of
‘GenerateTypeDependencies’ makes integer from pointer without a cast
[-Werror]
false);
^
In file included from pg_type.c:28:0:
../../../src/include/catalog/pg_type.h:335:13: note: expected ‘Oid’
but argument is of type ‘void *’
extern void GenerateTypeDependencies(Oid typeObjectId,
^
--
Thomas Munro
http://www.enterprisedb.com
On Mon, Nov 26, 2018 at 6:56 AM Tom Lane wrote:
> Thomas Munro writes:
> > Fix pushed.
> > By way of penance, I have now configured PG_TEST_EXTRA="ssl ldap
> > kerberos" for my build farm animals elver and eelpout. elver should
> > pass at the next build, a
3619:43: error: ‘outer_part’ may be used uninitialized in
this function [-Werror=maybe-uninitialized]
merged_index = map_and_merge_partitions(outer_pmap, outer_mmap,
^
partbounds.c:3475:6: note: ‘outer_part’ was declared here
int outer_part;
^
cc1: all warnings being treated as errors
--
Tho
ion makes integer from pointer without
a cast [-Werror]
},
^
guc.c:3328:3: error: (near initialization for
‘ConfigureNamesReal[19].gen.flags’) [-Werror]
--
Thomas Munro
http://www.enterprisedb.com
ith
> any configuration. This would require regenerating the certs in the
> tree. Any thoughts from others?
I don't really have opinion here, but I wanted to point out that
src/test/ldap/t/001_auth.pl creates new certs on the fly, which is a
bit inconsistent with the SSL test&
d here
other_sel,
^
extended_stats.c:1227:6: error: ‘mcv_sel’ may be used uninitialized in
this function [-Werror=maybe-uninitialized]
sel = mcv_sel + other_sel;
^
extended_stats.c:1087:5: note: ‘mcv_sel’ was declared here
mcv_sel,
^
--
Thomas Munr
On Fri, Nov 23, 2018 at 5:45 PM Thomas Munro
wrote:
> I do have a new plan though...
Ugh. The plan in my previous email doesn't work, I was confused about
the timing of the buffer header update. Back to the drawing board.
--
Thomas Munro
http://www.enterprisedb.com
tch cipher names with \w+, but some cipher names
sometimes (on some library versions?) have hyphens in them instead of
underscores.
https://travis-ci.org/postgresql-cfbot/postgresql/builds/459620336
--
Thomas Munro
http://www.enterprisedb.com
uld not find 7 free pages", which
should not happen. This is on 10.6, so it has the commit "Fix
segment_bins corruption in dsa.c.".
Hmm. I will see if I can come up with a many-partition torture test
reproducer for this.
[1]
https://www.postgresql.org/message-id/flat/CAJk1zg10iCNsxFvQ4pgKe1B0rdjNG9iELA7AzLXjXnQm5T%3DKzQ%40mail.gmail.com
--
Thomas Munro
http://www.enterprisedb.com
On Tue, Nov 27, 2018 at 4:00 PM Thomas Munro
wrote:
> Hmm. I will see if I can come up with a many-partition torture test
> reproducer for this.
No luck. I suppose one theory that could link both failure modes
would a buffer overrun, where in the non-shared case it trashes a
pointer t
On Tue, Nov 27, 2018 at 2:53 PM Haribabu Kommi wrote:
> Thanks for the updated patch. It looks good.
> I marked it in the commitfest as ready for committer.
Pushed. Thanks for the reviews!
--
Thomas Munro
http://www.enterprisedb.com
Hello,
Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce
it, but I noticed exactly the same thing on longfin the other day:
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=longfin&dt=2018-11-25%2005%3A39%3A04
Anyone know what that's about?
--
Tho
On Wed, Nov 28, 2018 at 4:10 PM Tom Lane wrote:
> Thomas Munro writes:
> > Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce
> > it, but I noticed exactly the same thing on longfin the other day:
> > https://buildfarm.postgresql.org/cgi-bin/show_log.p
On Wed, Nov 28, 2018 at 5:28 PM Tom Lane wrote:
> Thomas Munro writes:
> > Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce
> > it, but I noticed exactly the same thing on longfin the other day:
> > https://buildfarm.postgresql.org/cgi-bin/show_log.p
t;)
> pg_ctl: reads file, gets zero bytes
>
> But that idea is in the OS-bug class, which we agree seems unlikely.
I think you might be right.
https://github.com/macdice/unlinktest
--
Thomas Munro
http://www.enterprisedb.com
On Thu, Nov 29, 2018 at 3:30 AM Tom Lane wrote:
> Thomas Munro writes:
> > On Wed, Nov 28, 2018 at 6:47 PM Tom Lane wrote:
> >> Is it possible that unlink() on APFS is not atomic?
>
> > I think you might be right.
> > https://github.com/macdice/unlinktest
>
" binary.
Confirmed that if you pkg install python2 and python3 meta-packages on
a FreeBSD box you get python2 and python3 commands (symlinks to
python2.7, python3.6).
--
Thomas Munro
http://www.enterprisedb.com
On Thu, Nov 29, 2018 at 1:40 PM Ideriha, Takeshi
wrote:
> I found a minor typo in dsa.c.
>
> s/set_dsa_size_limit/dsa_set_size_limit/
Thanks! Pushed.
--
Thomas Munro
http://www.enterprisedb.com
DsT49RBaBJsaot_qNhPSCXBm9rzA%40mail.gmail.com
The relevant TL;DR is that if you're calling anything non-trivial in
the loop, you might get stuck in wait syscall if you use a half-baked
SIGTERM handler that only knows how to exit your top level loop, but
prevents CHECK_FOR_INTERRUPTS() from detecting shutdown.
--
Thomas Munro
http://www.enterprisedb.com
pass and show the feature working correctly. I think this
is getting close to committable. I see that Magnus has signed up as
committer.
--
Thomas Munro
http://www.enterprisedb.com
ls in the scan.
In other words, I wonder if our left-deep plans stand to gain more
from Bloom filter push down.
[1]
http://www.nus.edu.sg/nurop/2010/Proceedings/SoC/NUROP_Congress_Cheng%20Bin.pdf
[2]
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.73.8069&rep=rep1&type=pdf
[3]
Hi,
Here's a tiny patch to fix a typo.
--
Thomas Munro
http://www.enterprisedb.com
typo.patch
Description: Binary data
roup, the leader has gone away, so just exit quietly"
but for various other similar things we spew errors (most commonly
seen one being "ERROR: could not map dynamic shared memory segment").
Intentional?
--
Thomas Munro
http://www.enterprisedb.com
ssi-parallel-v11.patch
Description: Binary data
On Fri, Feb 23, 2018 at 1:54 AM, Thomas Munro
wrote:
> The attached is a draft patch only, needing some testing and polish.
> Brickbats, better ideas?
Note, that version is broken for multiple Gather nodes, but that's
fixable. Comments on the general idea welcome.
--
Thomas
On Fri, Feb 23, 2018 at 3:29 PM, Amit Kapila wrote:
> On Thu, Feb 22, 2018 at 10:35 PM, Robert Haas wrote:
>> On Thu, Feb 22, 2018 at 7:54 AM, Thomas Munro
>>> PS I noticed that for BecomeLockGroupMember() we say "If we can't
>>> join the lock group, t
On Fri, Feb 23, 2018 at 7:56 PM, Amit Kapila wrote:
> On Fri, Feb 23, 2018 at 8:48 AM, Thomas Munro
> wrote:
>> On Fri, Feb 23, 2018 at 3:29 PM, Amit Kapila wrote:
>>> By the way, in which case leader can exit early? As of now, we do
>>> wait for workers t
avept ... ok
Patch attached. I realise this isn't the most pressing problem faced
by computer science...
--
Thomas Munro
http://www.enterprisedb.com
test-name-inflation-adjustment.patch
Description: Binary data
On Fri, Feb 23, 2018 at 6:05 AM, Robert Haas wrote:
> On Thu, Feb 22, 2018 at 7:54 AM, Thomas Munro
> wrote:>
>> The best solution I have come up with so far is to add a reference
>> count to SERIALIZABLEXACT. I toyed with putting the refcount into the
>> DSM i
and no matter which
of its strategies it picked (either it receives a budget of work_mem /
participants, and then multiplies it by participants to create a
no-partition hash table combining the participants' budgets, or it
lets each participant chew on smaller hash tables adding up to at
Hi hackers,
Here is an attempt at updating parallel.sgml to cover Parallel Hash.
I will be neither surprised nor offended if Robert would like to put
it differently!
--
Thomas Munro
http://www.enterprisedb.com
update-parallel-join-docs.patch
Description: Binary data
On Sat, Feb 24, 2018 at 12:04 AM, Thomas Munro
wrote:
> I'm testing another version that is a lot simpler: like v10, it relies
> on the knowledge that the leader's transaction will always end after
> the workers have finished, but it handles the RO_SAFE optimisati
On Tue, Feb 13, 2018 at 7:47 PM, Thomas Munro
wrote:
> One way to avoid all that might be to use pseudo page numbers that
> don't suffer from splits. I don't know how you'd choose the
> constant, but it could be something like pseudo page number = hash
> value % 1024.
On Tue, Feb 27, 2018 at 4:18 AM, Tom Kazimiers wrote:
> On Mon, Feb 26, 2018 at 11:15:44PM +1300, Thomas Munro wrote:
>> On Sat, Feb 24, 2018 at 4:47 PM, Tom Kazimiers
>> wrote:
>> Thanks for the reproducer. Yeah, that seems to be a bug.
>> nodeNamedTuplestorescan.c a
27;;
+}
One thing that surprised me is that your test runs my system's
installed psql command out of my $PATH, not the one under test. Is
that expected?
--
Thomas Munro
http://www.enterprisedb.com
On Wed, Feb 28, 2018 at 8:39 AM, Robert Haas wrote:
> On Sat, Feb 10, 2018 at 6:07 PM, Thomas Munro
> wrote:
>> I forgot to register a display name for LWTRANCHE_PARALLEL_HASH_JOIN,
>> the tranche ID used by the LWLock that Parallel Hash uses when handing
>> out chunk
On Wed, Feb 28, 2018 at 9:58 AM, Tom Lane wrote:
> Thomas Munro writes:
>> Here's a new version with tuplestore_select_read_pointer() added in
>> another place where it was lacking, and commit message. Moving to
>> -hackers, where patches go.
>
> Pushed, along
On Wed, Jan 3, 2018 at 4:31 AM, Shubham Barai wrote:
> I have created new isolation tests. Please have a look at
> updated patch.
Hi Shubham,
Could we please have a rebased version of the gin one?
--
Thomas Munro
http://www.enterprisedb.com
GGUQ%40mail.gmail.com
[5]
https://www.postgresql.org/message-id/CAEepm%3D352uctNiFoN84UN4gtunbeTK-PBLouVe8i_b8ZPcJQFQ%40mail.gmail.com
--
Thomas Munro
http://www.enterprisedb.com
0001-Synchronous-replay-mode-for-avoiding-stale-reads--v5.patch
Description: Binary data
On Mon, Feb 26, 2018 at 6:37 PM, Thomas Munro
wrote:
> I've now broken it into two patches.
Rebased.
--
Thomas Munro
http://www.enterprisedb.com
0001-Enable-parallel-query-with-SERIALIZABLE-isolatio-v13.patch
Description: Binary data
0002-Enable-the-read-only-SERIALIZABLE-optimi
it, because it
currently prevents me from automatically testing all Commitfest
patches on the appveyor.com Windows build farm. I know that it passes
as a non-administrative user.
--
Thomas Munro
http://www.enterprisedb.com
On Thu, Mar 1, 2018 at 2:39 PM, Thomas Munro
wrote:
> I was pinged off-list by a fellow -hackers denizen interested in the
> synchronous replay feature and wanting a rebased patch to test. Here
> it goes, just in time for a Commitfest. Please skip to the bottom of
> this message
CB(crc32_c, *p_buf);
+}
>From the department of trivialities, our coding style has braces like this:
if (length & sizeof(uint16))
{
CRC32CH(crc32_c, *(uint16*)p_buf);
p_buf += sizeof(uint16);
}
if (length & sizeof(uint8))
CRC32CB(crc32_c, *p_buf);
--
Thomas Munro
http://www.enterprisedb.com
tion predicate locks at the cost of slightly higher collision rate
then we should consider that for the whole system (including heap page
predicate locking), not just hash indexes. Please consider those
ideas parked for now.
--
Thomas Munro
http://www.enterprisedb.com
ommitter?
>
> Done.
Thanks.
> Actually I think it'd be nice if the "parallel_hash" argument was mentioned in
> the header comment of initial_cost_hashjoin() function, but not sure this is
> worth returning the patch to the author.
Done.
--
Thomas Munro
http://www.enterprise
On Thu, Mar 1, 2018 at 4:43 AM, Peter Eisentraut
wrote:
> On 2/27/18 00:56, Thomas Munro wrote:
>> FWIW it passes for me if I add this:
>>
>> +elsif ($^O eq 'freebsd')
>> +{
>> + $krb5_bin_dir = '/usr/local/bin';
>> + $krb5_sb
1301 - 1400 of 4281 matches
Mail list logo
