Horton

Re: Non-reproducible AIO failure

Nico Williams <nico@cryptonector.com>

View thread
From: Nico Williams <nico@cryptonector.com>
To: Andres Freund <andres@anarazel.de>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Konstantin Knizhnik <knizhnik@garret.ru>, pgsql-hackers@lists.postgresql.org
Date: 2025-06-07T04:28:12Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. aio: Stop using enum bitfields due to bad code generation

  2. amcheck: Fix posting tree checks in gin_index_check()

  3. aio: Add missing memory barrier when waiting for IO handle

On Fri, Jun 06, 2025 at 03:37:45PM -0400, Andres Freund wrote:
> On 2025-06-06 15:21:13 -0400, Tom Lane wrote:
> > So it's our code that is busted.  No doubt, what is happening is
> > that process A is fetching two fields, modifying one of them,
> > and storing the word back (with the observed value of the other
> > field) concurrently with some other process trying to update
> > the other field.  So the other process's update is lost.
> 
> There shouldn't be any concurrent accesses here, so I don't really see
> how the above would explain the problem (the IO can only ever be
> modified by one backend, initially the "owning backend", then, when
> submitted, by the IO worker, and then again by the backend).

No concurrent accesses, not from threads anyways, but not even from
kernel subsystems nor from signal handlers?