Обсуждение: Shared Memory: How to use SYSV rather than MMAP ?

On Tue, Nov 20, 2018 at 11:11 PM REIX, Tony <tony.reix@atos.net> wrote:
> On AIX, since with MMAP we have only 4K pages though we can have 64K pages with SYSV, we'd like to experiment with
SYSVrather than MMAP and measure the impact to the performance.
 
>
> Looking at file: src/include/storage/dsm_impl.h , it seemed to me that replacing the line:
>
> #define DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE             DSM_IMPL_POSIX
> by the line:
> #define DEFAULT_DYNAMIC_SHARED_MEMORY_TYPE             DSM_IMPL_SYSV

Hi Tony,

SHOW dynamic_shared_memory_type to see which one it's actually using,
and set it in postgresql.conf to change it.

> However, when looking at details by means of procmap tool, it is unclear if that worked or not.

These segments are short-lived ones used for parallel query.  I
haven't used AIX recently but I suspect procmap -X will 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
page sizes due to better TLB cache hit ratios.

-- 
Thomas Munro
http://www.enterprisedb.com

On Tue, Nov 20, 2018 at 5:11 AM REIX, Tony <tony.reix@atos.net> wrote:
> On AIX, since with MMAP we have only 4K pages though we can have 64K pages with SYSV, we'd like to experiment with
SYSVrather than MMAP and measure the impact to the performance.
 

Are you trying to move the main shared memory segment or the dynamic
shared memory segments?

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On Tue, Nov 20, 2018 at 8:36 AM REIX, Tony <tony.reix@atos.net> wrote:
> We are trying to understand why pgbench on AIX is slower compared to Linux/Power on the same HW/Disks.
>
> So, we have yet no idea about what may be the root cause and what should be changed.
>
> So, changing:     dynamic_shared_memory_type = sysv    seems to help.
>
> And maybe changing the main shared memory segment could also improve the performance. However, how one can change
this?

There's no configuration setting for the main shared memory segment,
but removing #define USE_ANONYMOUS_SHMEM from sysv_shmem.c would
probably do the trick.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On Wed, Nov 21, 2018 at 4:37 AM REIX, Tony <tony.reix@atos.net> wrote:
> YES ! Reading this file, your suggestion should work ! Thx !
>
> I've rebuilt and run the basic tests. We'll relaunch our tests asap.

I would be surprised if that makes a difference:
anonymous-mmap-then-fork and SysV shm are just two different ways to
exchange mappings between processes, but I'd expect the virtual memory
object itself to be basically the same, in terms of constraints that
might affect page size at least.

If you were talking about mmap backed by a file (which is what you get
for temporary parallel query segments if you tell it to use
dynamic_shared_memory_type = mmap), that might be a different matter,
because then the block size of the file system backing it might come
into the picture and limit the kernel's options.  For example, that is
why (with default settings) Parallel Hash can'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

Hi,

On 2018-11-21 09:00:58 +1300, Thomas Munro wrote:
> On Wed, Nov 21, 2018 at 4:37 AM REIX, Tony <tony.reix@atos.net> wrote:
> > YES ! Reading this file, your suggestion should work ! Thx !
> >
> > I've rebuilt and run the basic tests. We'll relaunch our tests asap.
> 
> I would be surprised if that makes a difference:
> anonymous-mmap-then-fork and SysV shm are just two different ways to
> exchange mappings between processes, but I'd expect the virtual memory
> object itself to be basically the same, in terms of constraints that
> might affect page size at least.

I don't think that's true on many systems, FWIW. On linux there's
certainly different behaviour, and e.g. the way to get hugepages for
anon-mmap and SysV shmem aren't the same. [1] strongly suggests that
that's not the case on FreeBSD either (with sysv shmem being
better). I'd attached a patch to implement a GUC to allow users to
choose the shmem implementation back then [2].

[1] http://archives.postgresql.org/message-id/2AE143D2-87D3-4AD1-AC78-CE2258230C05%40FreeBSD.org
[2] http://archives.postgresql.org/message-id/20140422121921.GD4449%40awork2.anarazel.de

Greetings,

Andres Freund

On Wed, Nov 21, 2018 at 9:07 AM Andres Freund <andres@anarazel.de> wrote:
> On 2018-11-21 09:00:58 +1300, Thomas Munro wrote:
> > On Wed, Nov 21, 2018 at 4:37 AM REIX, Tony <tony.reix@atos.net> wrote:
> > > YES ! Reading this file, your suggestion should work ! Thx !
> > >
> > > I've rebuilt and run the basic tests. We'll relaunch our tests asap.
> >
> > I would be surprised if that makes a difference:
> > anonymous-mmap-then-fork and SysV shm are just two different ways to
> > exchange mappings between processes, but I'd expect the virtual memory
> > object itself to be basically the same, in terms of constraints that
> > might affect page size at least.
>
> I don't think that's true on many systems, FWIW. On linux there's
> certainly different behaviour, and e.g. the way to get hugepages for
> anon-mmap and SysV shmem aren't the same.

Right, when asking for them explicitly the API is different (SHM_HUGE
flag to shmget(), MAP_HUGETLB flag to mmap()).  Actually I was
expecting AIX to be more like FreeBSD and Solaris, where you don't do
that, the OS just decides what page size to give you, but after some
quality time with google I now see that it's more like Linux in the
SysV case... there is an explicit flag:

https://www.ibm.com/support/knowledgecenter/en/ssw_aix_71/com.ibm.aix.performance/large_pages_shared_mem_segs.htm

You also need some special privileges:

https://www.ibm.com/support/knowledgecenter/en/ssw_aix_71/com.ibm.aix.performance/large_page_ovw.htm

As for the main shared buffers area using anon-mmap, I wonder if it
would automagically use large pages if you have the privileges and set
the LDR_CNTRL environment variable (or the equivalent XCOFF header for
the binary):

https://www.ibm.com/support/knowledgecenter/en/ssw_aix_71/com.ibm.aix.performance/set_env_variable_lpages.htm

> [1] strongly suggests that
> that's not the case on FreeBSD either (with sysv shmem being
> better). I'd attached a patch to implement a GUC to allow users to
> choose the shmem implementation back then [2].

Surprising.  I'd like to know if that's still true.  SysV shm is not
nice, and if there is anything accidentally better about its
performance, I'd love to know what.  That report (slightly) predates
this work (maybe causally linked), which fixed various VM scale
problems hit by PostgreSQL:
http://www.kib.kiev.ua/kib/pgsql_perf_v2.0.pdf

-- 
Thomas Munro
http://www.enterprisedb.com

On Sat, Nov 24, 2018 at 4:54 AM REIX, Tony <tony.reix@atos.net> wrote:
> Here is a patch for enabling SystemV Shared Memory on AIX, for 64K or bigger page size, rather than using MMAP shared
memory,which is slower on AIX.
 

> We have tested this code with 64K pages and pgbench, on AIX 7.2 TL2 Power 9, and it provided a maximum of +37%
improvement.

You also mentioned changing from XLC to GCC.  Did you test the various
changes in isolation?  XLC->GCC, mmap->shmget, with/without
SHM_LGPAGE.  37% is a bigger performance change than I expected from
large pages, since reports from other architectures are single-digit
percentage increases with pgbench -S.

If just changing to GCC gives you a big speed-up, it could of course
just be different/better code generation (though that'd be a bit sad
for XLC), but I also wonder if the different atomics support in our
tree could be implicated.

> We'll test this code with Large Pages (SHM_LGPAGE | SHM_PIN | S_IRUSR | S_IWUSR flags of shmget() ) ASAP.
>
>
> However, I wanted first to get your comments about this change in order to improve it for acceptance.

I think we should respect the huge_pages GUC, as we do on Linux and
Windows (since there are downsides to using large pages, maybe not
everyone would want that).  It could even be useful to allow different
page sizes to be requested by GUC (I see that DB2 has an option to use
16GB pages -- yikes).  It also seems like a good idea to have a
shared_memory_type GUC as Andres proposed (see his link), instead of
using a compile time option.  I guess it was made a compile time
option because nobody could imagine wanting to go back to SysV shm!
(I'm still kinda surprised that MAP_ANONYMOUS memory can't be coaxed
into large pages by environment 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 Wed, Dec 19, 2018 at 4:17 AM REIX, Tony <tony.reix@atos.net> wrote:
> Here is the patch we are using now on AIX for enabling SysV shm for AIX, which improves greatly the performance on
AIX.
>
> It is compile time.
>
> It seems to me that you'd like this to become a shared_memory_type GUC. Correct? However, I do not know how to do.
>
>
> Even as-is, this patch would greatly improve the performance of PostgreSQL v11.1 in the field on AIX machines. So,
we'dlike this change to be available for AIX asap.
 
>
>
> What are the next steps to get this patch accepted? or What are your suggestions for improving it?

Hi Tony,

Since it's not fixing a bug, we wouldn't back-patch that into existing
releases.  But I agree that we should do something like this for
PostgreSQL 12, and I think we should make it user configurable.

Here is a quick rebase of Andres's shared_memory_type patch for
master, so that you can put shared_memory_type=sysv in postgresql.conf
to get the old pre-9.3 behaviour (this may also be useful for other
operating systems).  Here also is a "blind" patch that makes it
respect huge_pages=try/on on AIX (or at least, I think it does; I
don't have an AIX to try it, it almost certainly needs some
adjustments).  Thoughts?


--
Thomas Munro
http://www.enterprisedb.com

Thomas Munro <thomas.munro@enterprisedb.com> writes:
> Since it's not fixing a bug, we wouldn't back-patch that into existing
> releases.  But I agree that we should do something like this for
> PostgreSQL 12, and I think we should make it user configurable.

I'm -1 on making this user configurable via a GUC; that adds documentation
and compatibility burdens that we don't need, for something of no value
to 99.99% of users.  The fact that the default would need to be
platform-dependent just makes that tradeoff even worse.  I think the other
0.01% who need to change the default (and are bright enough to be doing
the right thing for the right reasons) could certainly handle something
like a pg_config_manual.h control symbol --- see USE_PPC_LWARX_MUTEX_HINT
for a precedent that I think applies well here.  So I'd favor just doing
it that way.

            regards, tom lane

On Wed, Dec 26, 2018 at 11:43 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Thomas Munro <thomas.munro@enterprisedb.com> writes:
> > Since it's not fixing a bug, we wouldn't back-patch that into existing
> > releases.  But I agree that we should do something like this for
> > PostgreSQL 12, and I think we should make it user configurable.
>
> I'm -1 on making this user configurable via a GUC; that adds documentation
> and compatibility burdens that we don't need, for something of no value
> to 99.99% of users.  The fact that the default would need to be
> platform-dependent just makes that tradeoff even worse.  I think the other
> 0.01% who need to change the default (and are bright enough to be doing
> the right thing for the right reasons) could certainly handle something
> like a pg_config_manual.h control symbol --- see USE_PPC_LWARX_MUTEX_HINT
> for a precedent that I think applies well here.  So I'd favor just doing
> it that way.

I disagree.  I think there is a growing body of evidence that
b0fc0df9364d2d2d17c0162cf3b8b59f6cb09f67 killed performance on many
types of non-Linux systems.  This is the first report I recall about
AIX, but there have been previous complaints about some BSD variants.

When I was working on developing that commit, I went and tried to find
out all of the different ways of getting some shared memory from
various operating systems and compared them.  Anonymous shared memory
allocated via mmap() was the hands-down winner in almost every
respect: supported on many systems, no annoying operating system
limits, automatic deallocation when the last process exits.  It had
the disadvantage that it didn't have an equivalent of nattch, which
meant that we had to keep a small System V segment around just for
that purpose, but otherwise it looked really good.

However, I only considered the situation from a functional point of
view. I never considered the possibility that the method used to
obtain shared memory from the operating system would affect the
performance of that shared memory. To my surprise, however, it does,
and on multiple operating systems from various parts of the UNIX
family tree. If I'd known that at the time, that commit probably would
not have gone into the tree in the form that it did.  I suspect that
there would have been a loud clamor for configurability, and I think I
would have agreed.

You may be right that this is of no value to a high percentage our
users, but I think that's only because a high percentage of our users
run Linux or Windows, which happen not to be affected. I'm rather
proud, though, of PostgreSQL's long history of trying to be
cross-platform. Even if operating systems like AIX or BSD are a small
percentage of the overall user base, I think it's totally fair to add
a GUC which likely be helpful to a large percentage of those people,
and I think the GUC proposed here likely falls into that category.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On December 26, 2018 6:48:31 PM GMT+01:00, Robert Haas <robertmhaas@gmail.com> wrote:
>I disagree.  I think there is a growing body of evidence that
>b0fc0df9364d2d2d17c0162cf3b8b59f6cb09f67 killed performance on many
>types of non-Linux systems.  This is the first report I recall about
>AIX, but there have been previous complaints about some BSD variants.

Exactly. I think we should have added this a few years ago.

Andres
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.

Robert Haas <robertmhaas@gmail.com> writes:
> On Wed, Dec 26, 2018 at 11:43 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> I'm -1 on making this user configurable via a GUC; that adds documentation
>> and compatibility burdens that we don't need, for something of no value
>> to 99.99% of users.
> ...
> You may be right that this is of no value to a high percentage our
> users, but I think that's only because a high percentage of our users
> run Linux or Windows, which happen not to be affected. I'm rather
> proud, though, of PostgreSQL's long history of trying to be
> cross-platform. Even if operating systems like AIX or BSD are a small
> percentage of the overall user base, I think it's totally fair to add
> a GUC which likely be helpful to a large percentage of those people,
> and I think the GUC proposed here likely falls into that category.

You misread what I said.  I don't say that we shouldn't fix this;
what I'm saying is we should not do so via a user-configurable knob.
We should be able to auto-configure this and just handle it internally.
I have zero faith in the idea that users would set the knob correctly.

            regards, tom lane

On Thu, Dec 27, 2018 at 6:48 AM Robert Haas <robertmhaas@gmail.com> wrote:
> On Wed, Dec 26, 2018 at 11:43 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> > Thomas Munro <thomas.munro@enterprisedb.com> writes:
> > > Since it's not fixing a bug, we wouldn't back-patch that into existing
> > > releases.  But I agree that we should do something like this for
> > > PostgreSQL 12, and I think we should make it user configurable.
> >
> > I'm -1 on making this user configurable via a GUC; that adds documentation
> > and compatibility burdens that we don't need, for something of no value
> > to 99.99% of users.  The fact that the default would need to be
> > platform-dependent just makes that tradeoff even worse.  I think the other
> > 0.01% who need to change the default (and are bright enough to be doing
> > the right thing for the right reasons) could certainly handle something
> > like a pg_config_manual.h control symbol --- see USE_PPC_LWARX_MUTEX_HINT
> > for a precedent that I think applies well here.  So I'd favor just doing
> > it that way.
>
> I disagree.  I think there is a growing body of evidence that
> b0fc0df9364d2d2d17c0162cf3b8b59f6cb09f67 killed performance on many
> types of non-Linux systems.  This is the first report I recall about
> AIX, but there have been previous complaints about some BSD variants.

FTR, I think the main FreeBSD complaint in a nutshell was:

1.  There are/were some lock contention problems on internal VM
objects and page tracking structures (pv entries copied for every
process sharing a VM object).
2.  SysV shmem offers a workaround in the form of sysctl
kern.ipc.shm_use_phys which effectively pins all SysV shared memory
(so it shows up in procstat -v as "ph" rather than "sw", ie it's not
swappable).  "Unmanaged pages never require finding all the instances
of their mappings, so the associated data structure used to find all
mappings (Described in Section 6.13) need not be allocated" (from 6.10
in the D&I of FreeBSD book).
3.  Shared anonymous mmap has no way to ask for that.

That wasn't all, though; see the PDF I mentioned up-thread.  I'm
hoping to learn more about this subject, hence my interest in reviving
that patch.  So far I can't reproduce the effect here, probably due to
lack of cores and probably also various changes that have been made
(but not the main ones described in that report, apparently).

-- 
Thomas Munro
http://www.enterprisedb.com

On Wed, Dec 26, 2018 at 1:57 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> You misread what I said.  I don't say that we shouldn't fix this;
> what I'm saying is we should not do so via a user-configurable knob.
> We should be able to auto-configure this and just handle it internally.
> I have zero faith in the idea that users would set the knob correctly.

I don't see how you can auto-configure a performance vs. usability
trade-off.  Remember, the original reason why we switched away from
doing a large System V shared memory allocation is that such
allocations tend to fail due to OS limits.  People who have those
limits set high enough will presumably prefer System V shared memory
allocation, while those who do not will prefer anonymous shared memory
over failure to start.  I guess we could try System V and fall back to
anonymous shared memory, but I think that's masterminding.  It risks
masking a performance problem that the user would like to notice and
fix.

Also, I doubt whether this is the only reason to have a GUC for this.
For instance, I seem to recall that there was some discussion of it
possibly being useful to let developers switch back to the
all-System-V approach for reasons that I don't recall at the moment,
and it even seems possible that somebody might want to use POSIX
shared memory so that they can open up the file that gets created and
inspect the contents using arbitrary tools.  I definitely agree that
an average user probably won't have any idea how to configure settings
like this, so we will want to think carefully about what the platform
defaults should be, but I also think that a GUC-less solution is
depriving those users who ARE smart enough to set the GUC the
opportunity to choose the value that works best for them.

Way back in high school somebody gave me a copy of the Camel book, and
it said something along the lines of: A good programming language
makes simple things simple and difficult things possible.  I thought
then, and still think now, that that's a very wise statement, and I
think it also applies to tools other than programming languages, like,
say, databases. Refusing to add a GUC is just deciding that we don't
trust our users with power tools, and that's not a philosophy with
which I can agree.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

On Thu, Dec 27, 2018 at 8:59 AM Thomas Munro
<thomas.munro@enterprisedb.com> wrote:
> ... So far I can't reproduce the effect here, probably due to
> lack of cores and probably also various changes that have been made
> (but not the main ones described in that report, apparently).

I decided to blow today's coffee money on a couple of hours on a 40
CPU m4.x10large at the Amazon arcade, running "FreeBSD
12.0-RELEASE-amd64-f5af2713-6d84-41f0-95ea-2eaee78f2af4-ami-03b0f822e17669866.4
(ami-0ec55ef66da2c9765)" on a 20GB SSD with UFS.  Some time when there
isn't a commitfest looming I'll try to do some proper testing, but in
a quick and dirty smoke test (-s 600, -S -T 60, shared_buffers=6GB,
for a range of client counts), mmap and sysv were the same, but there
did seem to be a measurable speed-up available at high client counts
with kern.ipc.shm_use_phys=1 and thus for sysv only, for people
prepared to set 3 sysctls and this proposed new GUC.  Maybe the effect
would be greater with bigger shared_buffers or smaller pages (this
test ran on super pages)?  More work required to figure that out.

Graph attached.

clients mmap    sysv    sysv+ph
======= ======= ======= =======
      1   13341   13277   13600
      3   39262   39431   39504
      6   78456   78702   78717
      9  116281  116356  116349
     12  149433  149284  149361
     15  169096  169224  169903
     18  177632  177774  178177
     24  202376  202574  203843
     32  227606  229967  231142
     48  291641  292431  302510
     64  287625  287118  298857
     80  277412  278449  293781

-- 
Thomas Munro
http://www.enterprisedb.com

On Thu, Dec 27, 2018 at 7:25 AM Andres Freund <andres@anarazel.de> wrote:
> On December 26, 2018 6:48:31 PM GMT+01:00, Robert Haas <robertmhaas@gmail.com> wrote:
> >I disagree.  I think there is a growing body of evidence that
> >b0fc0df9364d2d2d17c0162cf3b8b59f6cb09f67 killed performance on many
> >types of non-Linux systems.  This is the first report I recall about
> >AIX, but there have been previous complaints about some BSD variants.
>
> Exactly. I think we should have added this a few years ago.

I added a commitfest entry for this.

The 0001 patch (shared_memory_type GUC, code written by Andres, with a
first swing at documentation by me) seems close to committable, but as
Tom pointed out, there is a documentation burden here, and I'm
planning to go through and make sure that other relevant sections
explain the situation clearly.  I don't propose to change the default
on any platform.  It's a shame that it's still advantageous to use
clunky sysv facilities on some systems, but so long as that is the
case, it's good to be able to reach that; it's also good to have a
reasonable default that doesn't require any sysctl changes, so I think
this is a good thing to have as a GUC and mmap should be the default.

For the 0002 patch (essentially what Tony asked for, except made to
respect the huge_pages GUC by me, untested), I hope Tony or another
AIX user will be able to help get this into the right shape.  The main
idea there is that the default setting huge_pages=try should use large
pages if the OS is configured to support that, but otherwise fall back
to regular pages (just as we do on Linux and Windows).

-- 
Thomas Munro
http://www.enterprisedb.com

On 27/12/2018 00:53, Thomas Munro wrote:
> mmap and sysv were the same, but there
> did seem to be a measurable speed-up available at high client counts
> with kern.ipc.shm_use_phys=1 and thus for sysv only, for people
> prepared to set 3 sysctls and this proposed new GUC.  Maybe the effect
> would be greater with bigger shared_buffers or smaller pages (this
> test ran on super pages)?  More work required to figure that out.

Could you get a similar effect for mmap by using mlock() to prevent the
pages from being swapped?

-- 
Peter Eisentraut              http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Bonjour Tony,

On Sat, Jan 5, 2019 at 3:35 AM REIX, Tony <tony.reix@atos.net> wrote:
> Thanks for the 2 patches!
>
> I've seen also the discussion around this subject. Very interesting. Should I wait for a decision been taken? or
shouldI study and experiment your patches before?
 

I am planning to commit the 0001 patch shortly, unless there are
objections.  I attach a new version, which improves the documentation
a bit (cross-referencing the new GUC and the section on sysctl
settings).  That will give us shared_memory_type = sysv.

Can you please try out the 0002 patch on AIX and see if it works, and
if not, tell us how to fix it?  :-)  The desired behaviour is:

1.  If huge_pages = off, it doesn't use them.
2.  ff huge_pages = try, it tries to use them, but if it can't
(perhaps because the Unix user doesn't have CAP_BYPASS_RAC_VMM
capability), it should fall back to non-huge page.
3.  If huge_pages = on, it tries to use them, but if it can't it fails
to start up.

There may also be a case for supporting different page sizes
explicitly (huge_pages = 16GB?), but that could be done later.

-- 
Thomas Munro
http://www.enterprisedb.com

On Sat, Feb 2, 2019 at 2:23 AM REIX, Tony <tony.reix@atos.net> wrote:
> Hummmm I have an issue here:
>
> # cd /home2/freeware/src/packages/BUILD/postgresql-11.1/64bit/doc/src/sgml
> # gmake
> /opt/freeware/bin/xmllint --path . --noout --valid postgres.sgml
> /opt/freeware/bin/xsltproc --path . --stringparam pg.version '11.1'  stylesheet.xsl postgres.sgml
> Makefile:135: recipe for target 'html-stamp' failed
> gmake: *** [html-stamp] Segmentation fault (core dumped)
>
> That reaches an issue in  /opt/freeware/bin/xsltproc , on AIX.
> I never had this I think.

Strange.

By the way, the patches are for master (12-to-be), but I guess they'll
work fine on 11.x too.

> About testing it, I guess that I'll have to add:
>
>    huge_pages = on
>
> in the postgresql.conf file. Correct ?

Correct.

-- 
Thomas Munro
http://www.enterprisedb.com

On Sat, Feb 2, 2019 at 12:49 AM Thomas Munro
<thomas.munro@enterprisedb.com> wrote:
> I am planning to commit the 0001 patch shortly, unless there are
> objections.  I attach a new version, which improves the documentation
> a bit (cross-referencing the new GUC and the section on sysctl
> settings).  That will give us shared_memory_type = sysv.

Committed 0001.

I also tweaked the notes for both FreeBSD and NetBSD where we mention
kern.ipc.shm_use_phys to clarify that this refers only to System V
shared memory (before this commit, those notes were arguably a bit
misleading).

I noticed in passing that the comments about jails and System V shared
memory are out of date (the key namespace situation changed in FreeBSD
11), but I'll look into that another day.

I've moved the CF entry to the next Commitfest, since we still have to
fix up and commit the 0002 patch for AIX.

-- 
Thomas Munro
http://www.enterprisedb.com

On Sun, Feb 3, 2019 at 10:56 PM Thomas Munro
<thomas.munro@enterprisedb.com> wrote:
> Committed 0001.

> I've moved the CF entry to the next Commitfest, since we still have to
> fix up and commit the 0002 patch for AIX.

For the record, one problem with the shared_memory_type=sysv patch as
committed is that if you set huge_pages=on on Linux, it is allowed but
ignored.  I think we should respect it by passing SHM_HUGETLB to
shmget(), which, although not especially interesting as a feature
(given that there is no good reason for Linux users to prefer System V
shared memory anyway), it has the advantage that the code path would
be nearly identical to the proposed AIX huge page support (just a
different flag name), which is useful for development and testing (by
those of us with no AIX access).  Then the AIX support will be a very
tiny patch on top of that, which Tony can verify.

-- 
Thomas Munro
http://www.enterprisedb.com

On 2019-Feb-03, Thomas Munro wrote:

> On Sat, Feb 2, 2019 at 12:49 AM Thomas Munro
> <thomas.munro@enterprisedb.com> wrote:
> > I am planning to commit the 0001 patch shortly, unless there are
> > objections.  I attach a new version, which improves the documentation
> > a bit (cross-referencing the new GUC and the section on sysctl
> > settings).  That will give us shared_memory_type = sysv.
> 
> Committed 0001.

So can you please rebase what remains?

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

On Wed, Sep 4, 2019 at 10:30 AM Alvaro Herrera <alvherre@2ndquadrant.com> wrote:
> On 2019-Feb-03, Thomas Munro wrote:
> > On Sat, Feb 2, 2019 at 12:49 AM Thomas Munro
> > <thomas.munro@enterprisedb.com> wrote:
> > > I am planning to commit the 0001 patch shortly, unless there are
> > > objections.  I attach a new version, which improves the documentation
> > > a bit (cross-referencing the new GUC and the section on sysctl
> > > settings).  That will give us shared_memory_type = sysv.
> >
> > Committed 0001.
>
> So can you please rebase what remains?

Here's a quick rebase.  It needs testing, review and (probably)
adjustment from AIX users.  I'm not going to be able to do anything
with it on my own due to lack of access, though I'm happy to help get
this committed eventually.  If we don't get any traction in this CF,
I'll withdraw this submission for now.  For consistency, I think we
should eventually also do the same thing for Linux when using sysv
(it's pretty similar, it just uses different flag names; it may also
be necessary to query the page size and round up the requested size,
on one or both of those OSes; I'm not sure).

-- 
Thomas Munro
https://enterprisedb.com

On 2019-Sep-10, Thomas Munro wrote:

> Here's a quick rebase.  It needs testing, review and (probably)
> adjustment from AIX users.  I'm not going to be able to do anything
> with it on my own due to lack of access, though I'm happy to help get
> this committed eventually.  If we don't get any traction in this CF,
> I'll withdraw this submission for now.  For consistency, I think we
> should eventually also do the same thing for Linux when using sysv
> (it's pretty similar, it just uses different flag names; it may also
> be necessary to query the page size and round up the requested size,
> on one or both of those OSes; I'm not sure).

Tony, Sylvie, any chance for some testing on this patch?  It seems that
without that, this patch is going to waste.

If I don't hear from anyone on September 30, I'm going to close this as
Returned with Feedback.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Hi Tony,

On 2019-Sep-27, REIX, Tony wrote:

> Hello Thomas, Alvaro,
> 
> Sorry for the late answer, I missed your message of September 10. (I'm working on several different projects in
parallel.)
> Let me talk with Sylvie ASAP and see when I will be able to test it, probably next week, Tuesday. Is that OK for
you?

Sure -- I'm inclined to push this patch in state Needs Review to the
November commitfest in this case.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services