Hello,
RegisterExtensibleNodeMethods() initializes its hash table
with keysize=NAMEDATALEN, instead of EXTNODENAME_MAX_LEN.
The attached patch fixes it.
Thanks,
--
NEC Business Creation Division / PG-Strom Project
KaiGai Kohei
pgsql-v9.6-trivial-fix-extensiblenode.patch
Description: pgsql-v
Hello,
I'd like to adjust a few of custom-scan interface prior to v9.6 freeze.
The major point is serialization/deserialization mechanism.
Now, extension has to give LibraryName and SymbolName to reproduce
same CustomScanMethods on the background worker process side. Indeed,
it is sufficient info
ql.org; Amit Langote
> > Subject: Re: [HACKERS] Way to check whether a particular block is on the
> > shared_buffer?
> >
> > On Thu, Feb 11, 2016 at 9:05 PM, Kouhei Kaigai wrote:
> > > Hmm. In my experience, it is often not a productive discussion whether
&
Subject: Re: [HACKERS] Way to check whether a particular block is on the
> shared_buffer?
>
> On Thu, Mar 3, 2016 at 8:54 PM, Kouhei Kaigai wrote:
> > I found one other, but tiny, problem to implement SSD-to-GPU direct
> > data transfer feature under the PostgreSQL storage.
> On 29/02/16 13:07, Kouhei Kaigai wrote:
> >
> > I'd like to adjust a few of custom-scan interface prior to v9.6 freeze.
> >
> > The major point is serialization/deserialization mechanism.
> > Now, extension has to give LibraryName and SymbolName to repro
orks of CustomScan serialization/deserialization
>
> On 10/03/16 02:18, Kouhei Kaigai wrote:
> >
> >> I am not sure I like the fact that we have this EXTNODENAME_MAX_LEN and
> >> now the CUSTOM_NAME_MAX_LEN with the same length and also they are both
> >> same le
Hello,
I'm now checking the new planner implementation to find out the way to
integrate CustomPath to the upper planner also.
CustomPath node is originally expected to generate various kind of plan
node, not only scan/join, and its interface is designed to support them.
For example, we can expect
orks of CustomScan serialization/deserialization
>
> On 10/03/16 08:08, Kouhei Kaigai wrote:
> >>
> >>>> Also in RegisterCustomScanMethods
> >>>> +Assert(strlen(methods->CustomName) <= CUSTOM_NAME_MAX_LEN);
> >>>>
> >>>>
> On 14/03/16 02:53, Kouhei Kaigai wrote:
> >> -Original Message-
> >> From: pgsql-hackers-ow...@postgresql.org
> >> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Petr Jelinek
> >> Sent: Friday, March 11, 2016 12:27 AM
> >> To: Ka
gt;
> On 14/03/16 02:43, Kouhei Kaigai wrote:
> >
> > CustomPath node is originally expected to generate various kind of plan
> > node, not only scan/join, and its interface is designed to support them.
> > For example, we can expect a CustomPath that generates &quo
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Etsuro Fujita
> Sent: Monday, March 14, 2016 4:59 PM
> To: Ashutosh Bapat; Tom Lane
> Cc: pgsql-hackers
> Subject: Re: [HACKERS] Use %u to print user mapping's umid and
ion
>
> Petr Jelinek writes:
> > On 14/03/16 02:43, Kouhei Kaigai wrote:
> >> Even though I couldn't check the new planner implementation entirely,
> >> it seems to be the points below are good candidate to inject CustomPath
> >>
> On Sun, Mar 13, 2016 at 9:53 PM, Kouhei Kaigai wrote:
> > OK, I split the previous small patch into two tiny patches.
> > The one is bugfix around max length of the extnodename.
> > The other replaces Assert() by ereport() according to the upthread
> > discussion.
ubject: ##freemail## Re: [HACKERS] Reworks of CustomScan
> serialization/deserialization
>
> On Sun, Mar 13, 2016 at 9:53 PM, Kouhei Kaigai wrote:
> > OK, I split the previous small patch into two tiny patches.
> > The one is bugfix around max length of the extnodename.
>
.org
> Subject: Re: [HACKERS] WIP: Upper planner pathification
>
> Kouhei Kaigai writes:
> > On Wed, Mar 16, 2016 at 2:47 PM, Tom Lane wrote:
> >> I do not, however, like the proposal to expose wflists and so forth.
> >> Those are internal data structures in group
> Robert Haas writes:
> > On Mon, Mar 14, 2016 at 9:21 PM, Kouhei Kaigai wrote:
> >> So, even though we don't need to define multiple hook declarations,
> >> I think the hook invocation is needed just after create__paths()
> >> for each. It will nee
> > Robert Haas writes:
> > > On Mon, Mar 14, 2016 at 9:21 PM, Kouhei Kaigai
> > > wrote:
> > >> So, even though we don't need to define multiple hook declarations,
> > >> I think the hook invocation is needed just after create__paths()
> Robert Haas writes:
> > On Wed, Mar 16, 2016 at 2:47 PM, Tom Lane wrote:
> >> Robert Haas writes:
> >>> On Mon, Mar 14, 2016 at 9:21 PM, Kouhei Kaigai
> >>> wrote:
> >>>> So, even though we don't need to define multiple hook d
[HACKERS] Reworks of CustomScan serialization/deserialization
>
> On 15/03/16 05:03, Kouhei Kaigai wrote:
> > Petr,
> >
> > The attached patch is the revised one that follows the new extensible-
> > node routine.
> >
> > It is almost same the previous versio
> >> On 15/03/16 05:03, Kouhei Kaigai wrote:
> >>> Petr,
> >>>
> >>> The attached patch is the revised one that follows the new extensible-
> >>> node routine.
> >>>
> >>> It is almost same the previous version excep
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Kouhei Kaigai
> Sent: Saturday, March 19, 2016 8:57 AM
> To: Tom Lane
> Cc: Robert Haas; Petr Jelinek; David Rowley; pgsql-hackers@postgresql
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Robert Haas
> Sent: Friday, March 25, 2016 12:27 AM
> To: Petr Jelinek
> Cc: Kaigai Kouhei(海外 浩平); pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] Reworks of Cust
[HACKERS] Reworks of CustomScan serialization/deserialization
>
> On Mon, Mar 28, 2016 at 9:36 PM, Kouhei Kaigai wrote:
> > I don't have a strong reason to keep these stuff in separate files.
> > Both stuffs covers similar features and amount of code are enough small.
> &g
age-
> From: Kaigai Kouhei(海外 浩平)
> Sent: Wednesday, December 02, 2015 5:52 PM
> To: 'Robert Haas'
> Cc: Andres Freund; Amit Kapila; pgsql-hackers
> Subject: Re: CustomScan in a larger structure (RE: [HACKERS] CustomScan
> support
> on readfuncs.c)
>
&
> On Tue, Jan 19, 2016 at 7:59 AM, Greg Stark wrote:
> > On Mon, Jan 18, 2016 at 5:55 PM, Robert Haas wrote:
> >> For
> >> example, suppose that x and y are numeric columns and P(x) is
> >> length(x::text) == 3. Then you could have 1 in one table and 1.0 in
> >> the table; they join, but P(x) is
Hello,
What enhancement will be necessary to implement similar feature of
partial seq-scan using custom-scan interface?
It seems to me callbacks on the three points below are needed.
* ExecParallelEstimate
* ExecParallelInitializeDSM
* ExecParallelInitializeWorker
Anything else?
Does ForeignScan
Jan 26, 2016 at 12:00 PM, Kouhei Kaigai wrote:
> >
> > Hello,
> >
> > What enhancement will be necessary to implement similar feature of
> > partial seq-scan using custom-scan interface?
> >
> > It seems to me callbacks on the three points below are
> On Tue, Jan 26, 2016 at 1:30 AM, Kouhei Kaigai wrote:
> > What enhancement will be necessary to implement similar feature of
> > partial seq-scan using custom-scan interface?
> >
> > It seems to me callbacks on the three points below are needed.
'
> Cc: pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] CustomScan under the Gather node?
>
> > On Tue, Jan 26, 2016 at 1:30 AM, Kouhei Kaigai wrote:
> > > What enhancement will be necessary to implement similar feature of
> > > partial seq-scan using custo
> On Thu, Jan 28, 2016 at 10:50 AM, Kouhei Kaigai wrote:
> >> If I would make a proof-of-concept patch with interface itself, it
> >> seems to me file_fdw may be a good candidate for this enhancement.
> >> It is not a field for postgres_fdw.
> >>
> >
9.323 ms307.134 ms
Thanks,
--
NEC Business Creation Division / PG-Strom Project
KaiGai Kohei
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Kouhei Kaigai
> Sent: Friday, January 29, 2016 8:51 AM
>
> On Mon, Jan 25, 2016 at 8:06 PM, Kouhei Kaigai wrote:
> > Sorry for my late response. I've been unavailable to have enough
> > time to touch code for the last 1.5 month.
> >
> > The attached patch is a revised one to handle private data of
> > f
> On Thu, Jan 28, 2016 at 10:18 PM, Kouhei Kaigai wrote:
> > Do you think we shall allow to register same extensible node name for
> > different node types? Like, "GpuJoin" for any of CustomPath, CustomScan
> > and CustomScanState. Or, do we avoid this using differ
Hello,
Do we have a reliable way to check whether a particular heap block
is already on the shared buffer, but not modify?
Right now, ReadBuffer and ReadBufferExtended are entrypoint of the
buffer manager for extensions. However, it tries to acquire an
available buffer pool instead of the victim
> On 1/31/16 7:38 PM, Kouhei Kaigai wrote:
> > I'm under investigation of SSD-to-GPU direct feature on top of
> > the custom-scan interface. It intends to load a bunch of data
> > blocks on NVMe-SSD to GPU RAM using P2P DMA, prior to the data
> > loading onto CPU/R
> KaiGai-san,
>
> On 2016/02/01 10:38, Kouhei Kaigai wrote:
> > As an aside, background of my motivation is the slide below:
> > http://www.slideshare.net/kaigai/sqlgpussd-english
> > (LT slides in JPUG conference last Dec)
> >
> > I'm under investig
> > > On 1/31/16 7:38 PM, Kouhei Kaigai wrote:
>
> > > To answer your direct question, I'm no expert, but I haven't seen any
> > > functions that do exactly what you want. You'd have to pull relevant
> > > bits from ReadBuffer_*. Or maybe a be
Jan 28, 2016 at 8:14 PM, Kouhei Kaigai wrote:
> >> total ForeignScandiff
> >> 0 workers: 17584.319 ms 17555.904 ms 28.415 ms
> >> 1 workers: 18464.476 ms 18110.968 ms 353.508 ms
> >> 2 workers: 19042.755 ms 14580.335 m
> On Wed, Jan 27, 2016 at 9:36 PM, Robert Haas wrote:
> > On Mon, Jan 25, 2016 at 8:06 PM, Kouhei Kaigai wrote:
> >> Sorry for my late response. I've been unavailable to have enough
> >> time to touch code for the last 1.5 month.
> >>
> >> The
: Re: CustomScan in a larger structure (RE: [HACKERS] CustomScan
> support
> on readfuncs.c)
>
> On Wed, Feb 3, 2016 at 8:00 PM, Kouhei Kaigai wrote:
> >> Well, looking at this a bit more, it seems like the documentation
> >> you've written here is really mispl
> > KaiGai-san,
> >
> > On 2016/02/01 10:38, Kouhei Kaigai wrote:
> > > As an aside, background of my motivation is the slide below:
> > > http://www.slideshare.net/kaigai/sqlgpussd-english
> > > (LT slides in JPUG conference last Dec)
> > &
> On 2/4/16 12:30 AM, Kouhei Kaigai wrote:
> >> 2. A feature to suspend i/o write-out towards a particular blocks
> >> >that are registered by other concurrent backend, unless it is not
> >> >unregistered (usually, at the end of P2P DMA).
> >> &
ar block is on the
> shared_buffer?
>
> On 2/4/16 12:30 AM, Kouhei Kaigai wrote:
> >> 2. A feature to suspend i/o write-out towards a particular blocks
> >> >that are registered by other concurrent backend, unless it is not
> >> >unregistered (usually, at the
> On Wed, Feb 3, 2016 at 11:57 PM, Kouhei Kaigai wrote:
> > At this moment, I tried to write up description at nodes/nodes.h.
> > The amount of description is about 100lines. It is on a borderline
> > whether we split off this chunk into another header file, in my sense.
>
block is
> on the shared_buffer?
>
> On Thu, Feb 4, 2016 at 11:34 PM, Kouhei Kaigai wrote:
> > I can agree that smgr hooks shall be primarily designed to make storage
> > systems pluggable, even if we can use this hooks for suspend & resume of
> > write i/o stuff.
> > In a
ular block is
> on the shared_buffer?
>
> On Sun, Feb 7, 2016 at 9:49 PM, Kouhei Kaigai wrote:
> > On the other hands, it also became clear we have to guarantee OS buffer
> > or storage block must not be updated partially during the P2P DMA.
> > My motivation is a pote
CustomScan support on readfuncs.c)
>
> On Sun, Feb 7, 2016 at 7:28 PM, Kouhei Kaigai wrote:
> > The new callbacks of T_ExtensibleNode will replace the necessity to
> > form and deform process of private values, like as:
> > https://github.com/pg-strom/devel/blob/m
: Re: CustomScan in a larger structure (RE: [HACKERS] CustomScan
> support
> on readfuncs.c)
>
> On Wed, Feb 10, 2016 at 1:25 AM, Kouhei Kaigai wrote:
> > It is pretty good!
> >
> > The attached patch (primary one) implements the above idea.
> >
> >
> On Tue, Feb 9, 2016 at 6:35 PM, Kouhei Kaigai wrote:
> > Unfortunately, it was not sufficient.
> >
> > Due to the assumption, the buffer page to be suspended does not exist
> > when a backend process issues a series P2P DMA command. (If block would
> > be alread
gote
> Subject: Re: [HACKERS] Way to check whether a particular block is on the
> shared_buffer?
>
> On Thu, Feb 11, 2016 at 9:05 PM, Kouhei Kaigai wrote:
> > Hmm. In my experience, it is often not a productive discussion whether
> > a feature is niche or commodity. So, let me ch
Hello,
I noticed the source code comment around CustomPath structure says "see above"
for definition of CUSTOMPATH_* flags. It was originally right, but it was moved
to nodes/extensible.h on the further development. So, no comments are above.
The attached patch corrects the comment for the right l
Hello,
The attached patch adds an optional callback to support special optimization
if ForeignScan/CustomScan are located under the Limit node in plan-tree.
Our sort node wisely switches the behavior when we can preliminary know
exact number of rows to be produced, because all the Sort node has t
> On Mon, Aug 29, 2016 at 7:25 AM, Kouhei Kaigai wrote:
>
>
> Hello,
>
> The attached patch adds an optional callback to support special
> optimization
> if ForeignScan/CustomScan are located under the Limit node in plan-tree.
>
> Our
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Kouhei Kaigai
> Sent: Monday, September 05, 2016 12:58 PM
> To: Jeevan Chalke
> Cc: pgsql-hackers@postgresql.org; Etsuro Fujita
>
Sorry for my late.
The attached patch fixed the wording problems on SGML part.
Best regards,
--
NEC Business Creation Division / PG-Strom Project
KaiGai Kohei
> -Original Message-
> From: Jeevan Chalke [mailto:jeevan.cha...@enterprisedb.com]
> Sent: Tuesday, September 06, 2016 11:22 PM
> On Tue, Sep 13, 2016 at 3:48 AM, Kouhei Kaigai wrote:
> > Sorry for my late.
> >
> > The attached patch fixed the wording problems on SGML part.
>
> I agree that we should have some way for foreign data wrappers and
> custom scans and perhaps also other exec
Hello,
It looks to me pg_buffercache tries to allocate more than 1GB using
palloc(), when shared_buffers is more than 256GB.
# show shared_buffers ;
shared_buffers
280GB
(1 row)
# SELECT buffers, d.datname, coalesce(c.relname, '???')
FROM (SELECT count(*) buffers, reldatab
> On Wed, Sep 14, 2016 at 12:13 AM, Kouhei Kaigai wrote:
> > It looks to me pg_buffercache tries to allocate more than 1GB using
> > palloc(), when shared_buffers is more than 256GB.
> >
> > # show shared_buffers ;
> > shared_buffers
> >
Hello,
I'm now trying to carry extra performance statistics on CustomScan
(like DMA transfer rate, execution time of GPU kernels, etc...)
from parallel workers to the leader process using the DSM segment
attached by the parallel-context.
We can require an arbitrary length of DSM using ExecCustomSc
> I'm now trying to carry extra performance statistics on CustomScan
> (like DMA transfer rate, execution time of GPU kernels, etc...)
> from parallel workers to the leader process using the DSM segment
> attached by the parallel-context.
> We can require an arbitrary length of DSM using ExecCustom
> On Mon, Oct 17, 2016 at 6:22 AM, Kouhei Kaigai wrote:
> > Hello,
> >
> > I'm now trying to carry extra performance statistics on CustomScan
> > (like DMA transfer rate, execution time of GPU kernels, etc...)
> > from parallel workers to the leader process
gt; To: Kaigai Kouhei(海外 浩平)
> Cc: Robert Haas; pgsql-hackers
> Subject: ##freemail## Re: [HACKERS] Steps inside ExecEndGather
>
> On Mon, Oct 17, 2016 at 6:22 AM, Kouhei Kaigai wrote:
> > Hello,
> >
> > I'm now trying to carry extra performance statistics on
aigai Kouhei(海外 浩平)
> Cc: Jeevan Chalke; pgsql-hackers@postgresql.org; Etsuro Fujita; Andres Freund
> Subject: ##freemail## Re: [HACKERS] PassDownLimitBound for
> ForeignScan/CustomScan
>
> On Tue, Sep 13, 2016 at 9:07 PM, Kouhei Kaigai wrote:
> > In the current implementatio
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Dilip Kumar
> Sent: Saturday, October 29, 2016 3:48 PM
> To: Andres Freund
> Cc: Tom Lane; Alvaro Herrera; pgsql-hackers
> Subject: Re: [HACKERS] Proposal: scan key pus
ita; Andres Freund
> Subject: Re: [HACKERS] PassDownLimitBound for ForeignScan/CustomScan [take-2]
>
> On Mon, Oct 31, 2016 at 10:20 AM, Kouhei Kaigai wrote:
> > As an example, I enhanced postgres_fdw to understand the ps_numTuples
> > if it is set. If and when remote ORDER BY is
> On Tue, Nov 8, 2016 at 6:54 AM, Jeevan Chalke
> wrote:
> > 1. ps_numTuples is declared as long, however offset and count members in
> > LimitState struct and bound member in SortState struct is int64. However
> > long on 32 bit machine may be 32 bits and thus I think tuples_needed which
> > is
org; Jeevan Chalke
> ; Etsuro Fujita
> ; Andres Freund
> Subject: ##freemail## Re: PassDownLimitBound for ForeignScan/CustomScan
> [take-2]
>
> On Mon, Oct 31, 2016 at 10:20 AM, Kouhei Kaigai
> wrote:
> > As an example, I enhanced postgres_fdw to understand the ps_numTuple
sql-hackers-ow...@postgresql.org] On Behalf Of Claudio Freire
> Sent: Monday, February 06, 2017 3:37 PM
> To: Kaigai Kouhei(海外 浩平)
> Cc: Amit Kapila ; Robert Haas
> ; pgsql-hackers
> Subject: Re: ParallelFinish-hook of FDW/CSP (Re: [HACKERS] Steps inside
> ExecEndGather)
>
> On Mon,
ject: Re: ParallelFinish-hook of FDW/CSP (Re: [HACKERS] Steps inside
> ExecEndGather)
>
> On Fri, Feb 17, 2017 at 12:46 PM, Kouhei Kaigai
> wrote:
> > The attached patch is revised one.
> >
> > Invocation of Exec(Foreign|Custom)ParallelFinish was moved to
> > E
gt; Oops, I oversight this patch was marked as "returned with feedback", not
> "moved to the next CF".
>
> Its status has not been changed since the last update. (Code was revised
> according to the last comment by Jeevan, but CF-Nov was time up at that
> time.)
> Hello all,
>
> as this is my first mail to pgsql-hackers, please be gentle :)
>
Welcome to pgsql-hackers,
> I've looked at the patch, and as I'm not that familiar with the pg-sourcecode,
> customs and so on, this isn't a review, more like food for thought and all
> should be taken with a grain
> Hello,
>
> On Wed, March 1, 2017 7:21 pm, Kouhei Kaigai wrote:
> >> I've looked at the patch, and as I'm not that familiar with the
> >> pg-sourcecode, customs and so on, this isn't a review, more like food
> >> for thought and all should be t
n, and CustomPlanState became to
> >>> inherit ScanState. Because some useful routines to implement scan-
> >>> logic, like ExecScan, expects state-node has ScanState as its base
> >>> type, it's more kindness for extension side. (I'd like
Sorry for my late responding, now I'm catching up the discussion.
> * Robert Haas (robertmh...@gmail.com) wrote:
> > On Tue, Jul 1, 2014 at 3:20 PM, Dean Rasheed
> wrote:
> > > If RLS quals are instead regarded as constraints on access, and
> > > multiple policies apply, then it seems that the qu
> Kaigai,
>
> On Thursday, July 3, 2014, Kouhei Kaigai wrote:
>
>
> Sorry for my late responding, now I'm catching up the discussion.
>
> > * Robert Haas (robertmh...@gmail.com ) wrote:
> > > On Tue, Jul 1, 2014 at 3
> Alvaro Herrera writes:
> > I haven't followed this at all, but I just skimmed over it and noticed
> > the CustomPlanMarkPos thingy; apologies if this has been discussed
> > before. It seems a bit odd to me; why isn't it sufficient to have a
> > boolean flag in regular CustomPlan to indicate tha
: [HACKERS] [v9.5] Custom Plan API
>
> On 2014-07-16 10:43:08 +0900, Shigeru Hanada wrote:
> > Kaigai-san,
> >
> > 2014-07-15 21:37 GMT+09:00 Kouhei Kaigai :
> > > Sorry, expected result of sanity-check test was not updated on
> > > renaming to pg_cus
Fujita-san,
Sorry for my late response.
> The latest foreign-join pushdown patch allows fdw_scan_tlist to be set
> to a targetlist even for simple foreign table scans. However, since I
> think we assume that the test tuple of a foreign table for an EPQ
> testing, whether it may be copied from th
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Kyotaro HORIGUCHI
> Sent: Wednesday, July 22, 2015 4:10 PM
> To: robertmh...@gmail.com
> Cc: hlinn...@iki.fi; pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] Asyn
testing,
> doesn't it?
>
> Hi KaiGai-san,
>
> On 2015/07/22 16:44, Kouhei Kaigai wrote:
> >> The latest foreign-join pushdown patch allows fdw_scan_tlist to be set
> >> to a targetlist even for simple foreign table scans. However, since I
> >> think we a
Subject: Re: [HACKERS] fdw_scan_tlist for foreign table scans breaks EPQ
> testing,
> doesn't it?
>
> On Wed, Jul 22, 2015 at 8:13 AM, Kouhei Kaigai wrote:
> >> No, what I'm concerned about is the case when scanrelid > 0.
> >>
> > Hmm. if scanrelid > 0,
Hi Yang,
> I've performed some tests on pg_strom according to the wiki. But it seems that
> queries run slower on GPU than CPU. Can someone shed a light on what's wrong
> with my settings. My setup was Quadro K620 + CUDA 7.0 (For Ubuntu 14.10) +
> Ubuntu 15.04. And the results was
>
:
>
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Josh Berkus
> Sent: Thursday, July 23, 2015 2:49 AM
> To: YANG; pgsql-hackers@postgresql.org; KaiGai Kohei
> Subject: Re: [HACKERS] Queries runs slow on GPU with PG-Str
> > If we have ParallelAppend node that kicks a background worker process for
> > each underlying child node in parallel, does ForeignScan need to do
> > something
> > special?
>
> Although I don't see the point of the background worker in your
> story but at least for ParalleMergeAppend, it woul
Hi Amit,
The latest v16 patch cannot be applied to the latest
master as is.
434873806a9b1c0edd53c2a9df7c93a8ba021147 changed various
lines in heapam.c, so it probably conflicts with this.
[kaigai@magro sepgsql]$ cat ~/patch/parallel_seqscan_v16.patch | patch -p1
patching file src/backend/access/c
> On Wed, Jul 22, 2015 at 8:24 PM, Kouhei Kaigai wrote:
> > Indeed, this commit allows ForeignScan to have fdw_scan_tlist, even if
> > scanrelid > 0, however, I'm uncertain about its reason/intention.
> > Does it a preparation for the upcoming target-list-pushdown??
&
Hello Horiguchi-san,
> > > As for ForeignScan, it is merely an API for FDW and does nothing
> > > substantial so it would have nothing special to do. As for
> > > postgres_fdw, current patch restricts one execution per one
> > > foreign server at once by itself. We would have to provide
> > > anot
Fujita-san,
> On 2015/07/22 19:10, Etsuro Fujita wrote:
> > While working on the issue "Foreign join pushdown vs EvalPlanQual", I
> > happened to notice odd behaviors of late row locking in FDWs.
>
> > I think the reason for that is because we don't check pushed-down quals
> > inside an EPQ testi
ns
> breaks
> EPQ testing, doesn't it?
>
> On Thu, Jul 23, 2015 at 8:27 PM, Kouhei Kaigai wrote:
> > A dark side is, as discussed in this thread, complexity of EvalPlanQual.
> > RefetchForeignRow() returns a tuple based on foreign table definition,
> > on t
Hello,
I'm recently working/investigating on ParallelAppend feature
towards the next commit fest. Below is my design proposal.
1. Concept
--
Its concept is quite simple anybody might consider more than once.
ParallelAppend node kicks background worker process to execute
child nodes in par
Hello,
Under the investigation of ParallelAppend, I noticed here is a few
problems in CustomScan, that prevents to reproduce an equivalent
plan node on the background worker from serialized string.
1. CustomScanMethods->TextOutCustomScan callback
T
> Kouhei Kaigai writes:
> > Under the investigation of ParallelAppend, I noticed here is a few
> > problems in CustomScan, that prevents to reproduce an equivalent
> > plan node on the background worker from serialized string.
>
> > 1. CustomScanMetho
> On 04/01/2015 06:28 PM, Robert Haas wrote:
> > On Mon, Mar 30, 2015 at 1:28 AM, Michael Paquier
> > wrote:
> >>> I've been thinking of bumping this patch to the June commitfest as the
> >>> patch only exists to provide the basic infrastructure for things like
> >>> parallel aggregation, aggregat
> On 2015/07/24 23:51, Kouhei Kaigai wrote:
> >> On 2015/07/22 19:10, Etsuro Fujita wrote:
> >>> While working on the issue "Foreign join pushdown vs EvalPlanQual", I
> >>> happened to notice odd behaviors of late row locking in FDWs.
> >>
> Hello, can I ask some questions?
>
> I suppose we can take this as the analog of ParalleSeqScan. I
> can see not so distinction between Append(ParalleSeqScan) and
> ParallelAppend(SeqScan). What difference is there between them?
>
Append does not start to execute the second or later node until
f
> On Sun, Jul 26, 2015 at 8:43 AM, Kouhei Kaigai wrote:
> >
> > Hello,
> >
> > I'm recently working/investigating on ParallelAppend feature
> > towards the next commit fest. Below is my design proposal.
> >
> > 1. Concept
> > --
>
> -Original Message-
> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Kouhei Kaigai
> Sent: Monday, July 27, 2015 11:07 PM
> To: Amit Kapila
> Cc: pgsql-hackers@postgresql.org; Robert Haas; Kyotaro HORIGUCHI
>
uly 27, 2015 8:42 AM
> To: 'Tom Lane'
> Cc: pgsql-hackers@postgresql.org
> Subject: RE: [HACKERS] CustomScan and readfuncs.c
>
> > Kouhei Kaigai writes:
> > > Under the investigation of ParallelAppend, I noticed here is a few
> > > problems in CustomScan, t
> On Tue, Jul 28, 2015 at 7:59 AM, Kouhei Kaigai wrote:
> >
> > > -Original Message-
> > > From: pgsql-hackers-ow...@postgresql.org
> > > [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Kouhei Kaigai
> > > Sent: Monday,
> KaiGai-san,
>
> On 2015-07-27 PM 11:07, Kouhei Kaigai wrote:
> >
> > Append
> >--> Funnel
> > --> PartialSeqScan on rel1 (num_workers = 4)
> >--> Funnel
> > --> PartialSeqScan on rel2 (num_workers = 8)
1 - 100 of 457 matches
Mail list logo
