Re: [BUG] pg_stat_statements and extended query protocol

On Wed, Apr 05, 2023 at 05:39:13PM -0400, Tom Lane wrote:
> v5 seems OK to me except I think CreateExecutorState() should explicitly
> zero the new es_total_processed field, alongside zeroing es_processed.
> (I realize that the makeNode would have done it already, but our
> coding conventions generally run towards not relying on that.  This is
> mainly for greppability, so you can find where a field is initialized.)

Makes sense to me.  I'll look at that again today, potentially apply
the fix on HEAD.
--
Michael