From: Ronan Dunklau <ronan.dunklau@aiven.io>
To: Noah Misch <noah@leadboat.com>
Cc: Michael Paquier <michael@paquier.xyz>,
 pgsql-bugs <pgsql-bugs@lists.postgresql.org>
Subject: Re: FSM Corruption (was: Could not read block at end of the relation)
Date: Thu, 21 Mar 2024 14:51:25 +0100
Message-ID: <2493745.irdbgypaU6@aivenlaptop>
In-Reply-To: <20240316045834.83@rfd.leadboat.com>
References: 
 <VtPLBUerbMM8BSDLQwIMV9Co_38iHS1TXE4V09x2ZVFhPkZMNE8LwgPSIzB8PHkVFXykGx04QNAzZVjJFAnU1w==@protonmail.internalid>
 <20240316030339.03.nmisch@google.com> <20240316045834.83@rfd.leadboat.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="utf-8"
Archived-At: 
 <https://www.postgresql.org/message-id/2493745.irdbgypaU6%40aivenlaptop>
Precedence: bulk

Le samedi 16 mars 2024, 05:58:34 CET Noah Misch a =C3=A9crit :
> > Does "that benchmark" refer to "unlogged tables, 4 parallel copies",
> > "logged tables, 4 parallel copies", or something else?

Sorry I was quite busy, and will not find time to run more benchmarks until=
=20
next week but I'll get back to it.

It was 4 parallel copies, logged tables. Results for unlogged tables are=20
similar.=20

> >=20
> > > master: 15.05s
> > > patched: 15.24s (+1.25%)
> > >=20
> > > If I remove the best and worst run for each of those, the difference
> > > falls at +0.7%.
> >=20
> > To get some additional perspective on the benchmark, how hard would it =
be
> > to run one or both of the following?  Feel free to decline if difficult.
> >=20
> > - Make GetPageWithFreeSpace() just "return InvalidBlockNumber", then re=
run
> > the>=20
> >   benchmark.  This should be a lot slower.  If not, the bottleneck is
> >   somewhere unexpected, and we'd need a different benchmark.

Hum, it doesn't change much so that benchmark does not indeed exercise that=
=20
code too aggressively. I was thinking my initial benchmark for the very fir=
st=20
naive version of the patch could work, but I'm afraid it won't show much as=
=20
the table size will be cached soon enough.=20

> >=20
> > - Get profiles with both master and patched.  (lseek or freespace.c
> > functions>=20
> >   rising by 0.1%-1% would fit what we know.)
>=20

I'll try to run that asap, probably next week unfortunately.

> Forgot one more:
>=20
> - Your earlier version of the patch, with fewer lseek() but more disuse of
> FSM entries.

Noted.

=2D-
Ronan Dunklau