Re: [Patch] Optimize dropping of relation buffers using dlist - Mailing list pgsql-hackers

On Wed, Sep 23, 2020 at 7:56 AM tsunakawa.takay@fujitsu.com
<tsunakawa.takay@fujitsu.com> wrote:
>
> (3)
>         if (reln->smgr_cached_nblocks[forknum] == blocknum)
>                 reln->smgr_cached_nblocks[forknum] = blocknum + 1;
>         else
> +       {
> +               /*
> +                * DropRelFileNodeBuffers relies on the behavior that cached nblocks
> +                * won't be invalidated by file extension while recovering.
> +                */
> +               Assert(!InRecovery);
>                 reln->smgr_cached_nblocks[forknum] = InvalidBlockNumber;
> +       }
>
> I think this change is not directly related to this patch and can be a separate patch, but I want to leave the
decisionup to a committer.
 
>

We have added this mainly for testing purpose, basically this
assertion should not fail during the regression tests. We can keep it
in a separate patch but need to ensure that. If this fails then we
can't rely on the caching behaviour during recovery which is actually
required for the correctness of patch.


-- 
With Regards,
Amit Kapila.