Hello hackers,
I have a list of micro-optimisations and things to look into for hash
joins, which I've updated on the wiki[1]. Here's one that I was
inspired to poke at with a stick in a spare hour today.
Cache-oblivious hash joins cause a lot of TLB and cache misses.
Researchers tell us to redu
On Sat, Oct 13, 2018 at 09:45:33PM -0400, Tom Lane wrote:
> Notably, I ended up merging the two items about partitioned indexes and
> rewriting them altogether, because it seemed more sensible to present it
> that way. See what you think.
Looks good, thanks for commiting.
Justin
On Sat, Oct 13, 2018 at 05:53:00PM -0400, Andrew Dunstan wrote:
> It occurred to me that a pretty simple fix could just be to blacklist
> everything that didn't start with a digit. The whitelist approach is
> probably preferable... depends how urgent we see this as.
Yeah, possibly, still that's no
Justin Pryzby writes:
> On Sat, Oct 13, 2018 at 05:31:46PM -0400, Tom Lane wrote:
>> I've pushed updates merging Jonathan's last with my work. Please
>> look over what's in HEAD and see if you want to propose additional
>> changes.
> Find attached what's left.
I pushed some of this verbatim, tw
On Sat, Oct 13, 2018 at 05:31:46PM -0400, Tom Lane wrote:
> Justin Pryzby writes:
> >> On Sat, Oct 13, 2018 at 04:34:07PM -0400, Tom Lane wrote:
> >>> Hmm, I just saw this after spending an afternoon editing the release
> >>> notes. I'll try to merge it with what I did.
>
> > Feel free to send y
"Jung, Jinho" writes:
> Hello, I am Jinho Jung, Phd Student from GeorgiaTech, and try to find any SQL
> queries that cause performance regression. While conducting evaluation, I
> found an interesting query which makes x80 times slower execution in version
> 10.5 than version 9.4. Please see th
On 10/13/2018 10:00 AM, Andrew Dunstan wrote:
On 10/13/2018 04:30 AM, Michael Paquier wrote:
On Fri, Oct 12, 2018 at 12:14:48PM +0200, Michael Banck wrote:
On Fri, Oct 12, 2018 at 03:05:43PM +0900, Michael Paquier wrote:
On Fri, Oct 12, 2018 at 12:11:58PM +0900, Michael Paquier wrote:
Ag
Justin Pryzby writes:
> I suggest this should be backpatched ?
Yeah, done.
regards, tom lane
Justin Pryzby writes:
>> On Sat, Oct 13, 2018 at 04:34:07PM -0400, Tom Lane wrote:
>>> Hmm, I just saw this after spending an afternoon editing the release
>>> notes. I'll try to merge it with what I did.
> Feel free to send your own patch and I'll resolve conflicts and resend
> proposed
> chan
Hello, I am Jinho Jung, Phd Student from GeorgiaTech, and try to find any SQL
queries that cause performance regression. While conducting evaluation, I found
an interesting query which makes x80 times slower execution in version 10.5
than version 9.4. Please see the attached files, if you are i
I suggest this should be backpatched ?
commit 36e9d413a1d6928cd809d302d495fd6880a44b1e
Author: Greg Stark
Date: Sat Oct 6 13:03:43 2018 -0400
Add "B" suffix for bytes to docs
6e7baa3227 and b06d8e58b5 added "B" as a valid suffix for
GUC_UNIT_BYTES but neglected to add it to th
On Sat, Oct 13, 2018 at 03:38:39PM -0500, Justin Pryzby wrote:
> On Sat, Oct 13, 2018 at 04:34:07PM -0400, Tom Lane wrote:
> > "Jonathan S. Katz" writes:
> > > On 10/6/18 9:42 AM, Justin Pryzby wrote:
> > >> Find below various fixes to release notes for v11, for discussion
> > >> purposes.
> >
>
On 10/13/18 4:38 PM, Justin Pryzby wrote:
> On Sat, Oct 13, 2018 at 04:34:07PM -0400, Tom Lane wrote:
>> "Jonathan S. Katz" writes:
>>> On 10/6/18 9:42 AM, Justin Pryzby wrote:
Find below various fixes to release notes for v11, for discussion purposes.
>>
>>> Thanks for putting this together.
On Sat, Oct 13, 2018 at 04:34:07PM -0400, Tom Lane wrote:
> "Jonathan S. Katz" writes:
> > On 10/6/18 9:42 AM, Justin Pryzby wrote:
> >> Find below various fixes to release notes for v11, for discussion purposes.
>
> > Thanks for putting this together. I reviewed, broke it up into two
> > patches
"Jonathan S. Katz" writes:
> On 10/6/18 9:42 AM, Justin Pryzby wrote:
>> Find below various fixes to release notes for v11, for discussion purposes.
> Thanks for putting this together. I reviewed, broke it up into two
> patches, and made some additions / changes / deletions. Please see
> attached
Hi Justin,
On 10/6/18 9:42 AM, Justin Pryzby wrote:
> Find below various fixes to release notes for v11, for discussion purposes.
Thanks for putting this together. I reviewed, broke it up into two
patches, and made some additions / changes / deletions. Please see
attached patches.
Thanks!
Jonat
Attached is v6, a simple rebase.
-John Naylor
From e37cb80ab8e7baaa5231fc3b8dbc9d96ec253018 Mon Sep 17 00:00:00 2001
From: John Naylor
Date: Sat, 13 Oct 2018 19:28:08 +0700
Subject: [PATCH v6 1/2] Add pg_language lookup.
This didn't seem worth doing before, but an upcoming commit will add
88 ent
On 10/13/18, Amit Kapila wrote:
> On Sat, Oct 6, 2018 at 12:17 AM John Naylor wrote:
>> -For normal mode, I taught fsm_set_and_search() to switch to a
>> non-extending buffer call, but the biggest missing piece is WAL
>> replay.
>>
>
> fsm_set_and_search()
> {
> ..
> + /*
> + * For heaps we preve
On Thu, Oct 4, 2018 at 04:05:17PM -0300, EmÃlio B. Pedrollo wrote:
> Today Brazil's president announced that the DST for all our time zones would
> start at November 18th instead of November 4th
> There's already a thread at IANA about it
> https://mm.icann.org/pipermail/tz/2018-October/026921.htm
Hello Kirk,
I have decided to take a look into this patch.
Thanks for the feedback.
I noticed in your refactored code that CSTATE_CHOOSE_SCRIPT may now also
change to CSTATE_FINISHED when timer is exceeded. (Before, it only
changes to either CSTATE_START_THROTTLE or CSTATE_START_TX) But th
Hello Andrew,
A question: would it makes sense to have a symmetrical
--include-database=PATTERN option as well?
I don't think so. If you only want a few databases, just use pg_dump. The
premise of pg_dumpall is that you want all of them and this switch provides
for exceptions to that.
Ok
On 10/13/2018 04:30 AM, Michael Paquier wrote:
On Fri, Oct 12, 2018 at 12:14:48PM +0200, Michael Banck wrote:
On Fri, Oct 12, 2018 at 03:05:43PM +0900, Michael Paquier wrote:
On Fri, Oct 12, 2018 at 12:11:58PM +0900, Michael Paquier wrote:
Agreed. I am just working on a patch for v11- whic
On Sat, Oct 6, 2018 at 12:17 AM John Naylor wrote:
>
> Hi all,
> A while back, Robert Haas noticed that the space taken up by very
> small tables is dominated by the FSM [1]. Tom suggested that we could
> prevent creation of the FSM until the heap has reached a certain
> threshold size [2]. Attach
On Sat, 13 Oct 2018 at 02:02, Tom Lane wrote:
> Sure, but even a generated security token seems unlikely to be more
> than a couple dozen bytes long. What's the actual use-case for tokens
> longer than that? ISTM that a limit around 100 bytes already has a
> whole lot of headroom.
Self-contain
On Thu, Oct 11, 2018 at 5:51 PM Tom Lane wrote:
> The comment adjacent to the change in InitStandaloneProcess bothers me.
> In particular, it points out that what BackendRun() is currently doing
> creates more entropy in the process's random seed than what you have
> here: MyStartTime is only good
On Fri, Oct 12, 2018 at 12:14:48PM +0200, Michael Banck wrote:
> On Fri, Oct 12, 2018 at 03:05:43PM +0900, Michael Paquier wrote:
>> On Fri, Oct 12, 2018 at 12:11:58PM +0900, Michael Paquier wrote:
>>> Agreed. I am just working on a patch for v11- which uses a
>>> whitelist-based method instead of
26 matches
Mail list logo
