Re: index prefetching

Hi,

On 2025-07-16 17:27:23 -0400, Peter Geoghegan wrote:
> On Wed, Jul 16, 2025 at 4:46 PM Andres Freund <andres(at)anarazel(dot)de> wrote:
> > Maybe I'm missing something, but the current interface doesn't seem to work
> > for AMs that don't have a 1:1 mapping between the block number portion of the
> > tid and the actual block number?
>
> I'm not completely sure what you mean here.
>
> Even within nbtree, posting list tuples work by setting the
> INDEX_ALT_TID_MASK index tuple header bit. That makes nbtree interpret
> IndexTupleData.t_tid as metadata (in this case describing a posting
> list). Obviously, that isn't "a standard IndexTuple", but that won't
> break either patch/approach.
>
> The index AM is obligated to pass back heap TIDs, without any external
> code needing to understand these sorts of implementation details. The
> on-disk representation of TIDs remains an implementation detail known
> only to index AMs.

I don't mean the index tids, but how the read stream is fed block numbers. In
the "complex" patch that's done by index_scan_stream_read_next(). And the
block number it returns is simply

return ItemPointerGetBlockNumber(tid);

without the table AM having any way of influencing that. Which means that if
your table AM does not use the block number of the tid 1:1 as the real block
number, the fetched block will be completely bogus.

It's similar in the simple patch, bt_stream_read_next() etc also just use
ItemPointerGetBlockNumber().

Greetings,

Andres Freund