Horton

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

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: "Guo, Adam" <adamguo@amazon.com>
Cc: "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
Date: 2024-04-23T14:57:36Z
Lists: pgsql-hackers 
"Guo, Adam" <adamguo@amazon.com> writes:
> I would like to report an issue with the pg_trgm extension on
> cross-architecture replication scenarios. When an x86_64 standby
> server is replicating from an aarch64 primary server or vice versa,
> the gist_trgm_ops opclass returns different results on the primary
> and standby.

I do not think that is a supported scenario.  Hash functions and
suchlike are not guaranteed to produce the same results on different
CPU architectures.  As a quick example, I get

regression=# select hashfloat8(34);
 hashfloat8 
------------
   21570837
(1 row)

on x86_64 but

postgres=# select hashfloat8(34);
 hashfloat8 
------------
 -602898821
(1 row)

on ppc32 thanks to the endianness difference.

> Given that this has problem has come up before and seems likely to
> come up again, I'm curious what other broad solutions there might be
> to resolve it?

Reject as not a bug.  Discourage people from thinking that physical
replication will work across architectures.

			regards, tom lane

Commits

  1. pg_upgrade: Check for the expected error message in TAP tests.

  2. Fix a typo in 005_char_signedness.pl test.

  3. Add test 005_char_signedness.pl to meson.build.

  4. Fix an issue with index scan using pg_trgm due to char signedness on different architectures.

  5. pg_upgrade: Add --set-char-signedness to set the default char signedness of new cluster.

  6. pg_upgrade: Preserve default char signedness value from old cluster.

  7. pg_resetwal: Add --char-signedness option to change the default char signedness.

  8. Add default_char_signedness field to ControlFileData.

  9. Remove unneeded nbtree array preprocessing assert.