Re: pg_trgm comparison bug on cross-architecture replication due to different char implementation

From: Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
To: Noah Misch <noah(at)leadboat(dot)com>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Joe Conway <mail(at)joeconway(dot)com>, Peter Eisentraut <peter(at)eisentraut(dot)org>, Alexander Korotkov <aekorotkov(at)gmail(dot)com>, "Guo, Adam" <adamguo(at)amazon(dot)com>, "pgsql-hackers(at)lists(dot)postgresql(dot)org" <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Nathan Bossart <nathandbossart(at)gmail(dot)com>, Jim Mlodgenski <jimmy76(at)gmail(dot)com>
Subject: Re: pg_trgm comparison bug on cross-architecture replication due to different char implementation
Date: 2025-01-18 01:11:41
Message-ID: CAD21AoBGokmQPGPUnk_p8O2OW5fzeDLF+QYD5b55kH4k80-kRg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

I apologize for the late response. I completely missed tracking
updates on this thread.

On Tue, Nov 19, 2024 at 5:28 PM Noah Misch <noah(at)leadboat(dot)com> wrote:
>
> On Thu, Oct 03, 2024 at 06:55:47AM -0700, Masahiko Sawada wrote:
> > I've attached PoC patches for the idea Noah proposed. Newly created
> > clusters unconditionally have default_char_signedness=true, and the
> > only source of signedness=false is pg_upgrade. To update the
> > signedness in the controlfile, pg_resetwal now has a new option
> > --char-signedness, which is used by pg_upgrade internally. Feedback is
> > very welcome.
>
> Upthread, we discussed testability. Does pg_resetwal facilitate all
> appropriate testing, or do testing difficulties remain?

I think the new pg_resetwal facility to change the cluster's char
signedness is useful for testing. We can easily test cross
architecture replication from now on.

>
> I reviewed these patches, finding only one non-cosmetic review comment. Given
> the PoC status, some of the observations below are likely ones you already
> know or would have found before exiting PoC.

Thank you for reviewing the patches. I agree with all the comments you
made. I've addressed them and I've attached new version patches that
now have some regression tests.

Regards,

--
Masahiko Sawada
Amazon Web Services: https://aws.amazon.com

Attachment Content-Type Size
v3-0004-pg_upgrade-Add-set-char-signedness-to-set-the-def.patch application/octet-stream 8.1 KB
v3-0005-Fix-an-issue-with-index-scan-using-pg_trgm-due-to.patch application/octet-stream 4.3 KB
v3-0003-pg_upgrade-Preserve-default-char-signedness-value.patch application/octet-stream 8.6 KB
v3-0002-pg_resetwal-Add-char-signedness-option-to-change-.patch application/octet-stream 4.9 KB
v3-0001-Add-default_char_signedness-field-to-ControlFileD.patch application/octet-stream 8.7 KB

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Junwang Zhao 2025-01-18 01:23:19 rename es_epq_active to es_epqstate
Previous Message Jeff Davis 2025-01-18 00:40:07 Re: Unicode full case mapping: PG_UNICODE_FAST, and standard-compliant UCS_BASIC