Re: pg15b1: FailedAssertion("val > base", File: "...src/include/utils/relptr.h", Line: 67, PID: 30485)

Robert Haas <robertmhaas@gmail.com>

From: Robert Haas <robertmhaas@gmail.com>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Kyotaro Horiguchi <horikyota.ntt@gmail.com>, Justin Pryzby <pryzby@telsasoft.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-05-31T21:01:04Z
Lists: pgsql-hackers
On Tue, May 31, 2022 at 4:32 PM Thomas Munro <thomas.munro@gmail.com> wrote:
> This FPM isn't in a DSM.  (It happens to have DSMs *inside it*,
> because I'm using it as a separate DSM allocator: instead of making
> them with dsm_impl.c mechanisms, this one recycles space from the main
> shmem area).  I view FPM as a reusable 4kb page-based memory allocator
> that could have many potential uses, not as a thing that must live
> inside another thing with a TOC.  The fact that it uses the relptr
> thing makes it possible to use FPM inside DSMs too, but that doesn't
> mean it has to be used inside a DSM.

Could it use something other than its own address as the base address?
One way to do this would be to put it at the *end* of the
"Preallocated DSM" space, rather than the beginning.

-- 
Robert Haas
EDB: http://www.enterprisedb.com



Commits

  1. Fix relptr's encoding of the base address.

  2. Suppress compiler warning in relptr_store().