Re: Patch to address creation of PgStat* contexts with null parent context

At Thu, 28 Jul 2022 22:03:13 +0800, Zhang Mingli <zmlpostgres(at)gmail(dot)com> wrote in
> Hi,
>
> On Jul 28, 2022, 21:30 +0800, Reid Thompson <reid(dot)thompson(at)crunchydata(dot)com>, wrote:
> > Attached is a patch to address this.

Good Catch!

> Codes seem good, my question is:
>
> Do auto vacuum processes need CacheMemoryContext?

pgstat_report_vacuum requires it. Startup process doesn't seem to use
pgstats while recovery proceeding but requires the context only at
termination...

> Is it designed not to  create CacheMemoryContext in such processes?
>
> If so, we’d better use TopMemoryContext in such processes.

That makes the memorycontext-tree structure unstable because
CacheMemoryContext can be created on-the-fly.

Honestly I don't like to call CreateCacheMemoryContext in the two
functions on-the-fly. Since every process that calls
pgstat_initialize() necessarily calls pgstat_setup_memcxt() at latest
at process termination, I think we can create at least
CacheMemoryContext in pgstat_initialize(). Or couldn't we create the
all three contexts in the function, instead of calling
pgstat_setup_memcxt() on-the fly?

regards.

--
Kyotaro Horiguchi
NTT Open Source Software Center