Re: Hybrid Hash/Nested Loop joins and caching results from subplans

On Thu, 1 Apr 2021 at 23:41, houzj.fnst@fujitsu.com
<houzj.fnst@fujitsu.com> wrote:
>
> > I've attached the updated patch.  I'll let the CFbot grab this to ensure it's
> > happy with it before I go looking to push it again.
>
> Hi,
>
> I took a look into the patch and noticed some minor things.
>
> 1.
> +               case T_ResultCache:
> +                       ptype = "ResultCache";
> +                       subpath = ((ResultCachePath *) path)->subpath;
> +                       break;
>                 case T_UniquePath:
>                         ptype = "Unique";
>                         subpath = ((UniquePath *) path)->subpath;
> should we use "case T_ResultCachePath" here?
>
> 2.
> Is it better to add ResultCache's info to " src/backend/optimizer/README " ?
> Something like:
>   NestPath      - nested-loop joins
>   MergePath     - merge joins
>   HashPath      - hash joins
> + ResultCachePath - Result cache

Thanks for pointing those two things out.

I've pushed the patch again with some updates to EXPLAIN to fix the
issue from yesterday.  I also disabled result cache in the
partition_prune tests as I suspect that the parallel tests there might
just be a bit too unstable in the buildfarm. The cache
hit/miss/eviction line might disappear if the main process does not
get a chance to do any work.

Well, it's now time to watch the buildfarm again...

David