Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qhLyv-002AvX-0S for pgsql-performance@arkaria.postgresql.org; Sat, 16 Sep 2023 03:28:29 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.94.2) (envelope-from ) id 1qhLxt-005AT9-C5 for pgsql-performance@arkaria.postgresql.org; Sat, 16 Sep 2023 03:27:25 +0000 Received: from makus.postgresql.org ([2001:4800:3e1:1::229]) by malur.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qhLxs-005ASc-Sw for pgsql-performance@lists.postgresql.org; Sat, 16 Sep 2023 03:27:24 +0000 Received: from cloud.gatewaynet.com ([185.90.37.94]) by makus.postgresql.org with esmtps (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1qhLxp-004pOA-AF for pgsql-performance@lists.postgresql.org; Sat, 16 Sep 2023 03:27:23 +0000 Content-Type: multipart/alternative; boundary="------------7dEyXLApqd0iks1FwMvyvu3O" Message-ID: <2c350d20-0c1f-b9f3-4c65-73b07d8e3f40@cloud.gatewaynet.com> Date: Sat, 16 Sep 2023 06:27:16 +0300 MIME-Version: 1.0 Subject: Re: pgsql 10.23 , different systems, same table , same plan, different Buffers: shared hit Content-Language: en-US To: Thomas Munro , Tom Lane Cc: pgsql-performance@lists.postgresql.org References: <24a36ca8-b5c0-d39f-c5f6-47cf2ef51abd@cloud.gatewaynet.com> <3310210.1694791429@sss.pgh.pa.us> <3666541.1694806930@sss.pgh.pa.us> From: Achilleas Mantzios In-Reply-To: List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Archived-At: Precedence: bulk This is a multi-part message in MIME format. --------------7dEyXLApqd0iks1FwMvyvu3O Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Στις 16/9/23 02:08, ο/η Thomas Munro έγραψε: > On Sat, Sep 16, 2023 at 7:42 AM Tom Lane wrote: >> Sadly, this proves very little about Linux's behavior. glibc's idea >> of en_US involves some very complicated multi-pass sort rules. >> AFAICT from the FreeBSD sort(1) man page, FreeBSD defines en_US >> as "same as C except case-insensitive", whereas I'm pretty sure >> that underscores and other punctuation are nearly ignored in >> glibc's interpretation; they'll only be taken into account if the >> alphanumeric parts of the strings sort equal. > Achilleas didn't mention the glibc version, but based on the kernel > vintage mentioned I guess that must be the "old" (pre 2.28) glibc > sorting. In 2.28 they did a big sync-up with ISO 14651, while FreeBSD > follows the UCA, a closely related standard[1]. I think newer > Linux/glibc systems should agree with FreeBSD's libc in more cases > (and also agree with ICU). Thank you Thomas , our linux's glibc is on version : 2.19-18+deb8u10, we need to upgrade on so many levels. > > [1]https://unicode.org/reports/tr10/#Synch_ISO14651 -- Achilleas Mantzios IT DEV - HEAD IT DEPT Dynacom Tankers Mgmt --------------7dEyXLApqd0iks1FwMvyvu3O Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 8bit
Στις 16/9/23 02:08, ο/η Thomas Munro έγραψε:
On Sat, Sep 16, 2023 at 7:42 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:

Sadly, this proves very little about Linux's behavior.  glibc's idea
of en_US involves some very complicated multi-pass sort rules.
AFAICT from the FreeBSD sort(1) man page, FreeBSD defines en_US
as "same as C except case-insensitive", whereas I'm pretty sure
that underscores and other punctuation are nearly ignored in
glibc's interpretation; they'll only be taken into account if the
alphanumeric parts of the strings sort equal.

Achilleas didn't mention the glibc version, but based on the kernel
vintage mentioned I guess that must be the "old" (pre 2.28) glibc
sorting.  In 2.28 they did a big sync-up with ISO 14651, while FreeBSD
follows the UCA, a closely related standard[1].  I think newer
Linux/glibc systems should agree with FreeBSD's libc in more cases
(and also agree with ICU).
Thank you Thomas , our linux's glibc is on version : 2.19-18+deb8u10, we need to upgrade on so many levels.

[1] https://unicode.org/reports/tr10/#Synch_ISO14651

-- 
Achilleas Mantzios
 IT DEV - HEAD
 IT DEPT
 Dynacom Tankers Mgmt
--------------7dEyXLApqd0iks1FwMvyvu3O--