wenjing zeng <wjzeng2...@gmail.com> writes:
> I found that in the index_update_stats function, i.e. the CREATE
> INDEX/REINDEX/Truncate INDEX process,
> relchche is invalidated whether the index information is updated. I want to
> know why you're did this
Did you read the function's header comment? It says
* NOTE: an important side-effect of this operation is that an SI invalidation
* message is sent out to all backends --- including me --- causing relcache
* entries to be flushed or updated with the new data. This must happen even
* if we find that no change is needed in the pg_class row. When updating
* a heap entry, this ensures that other backends find out about the new
* index. When updating an index, it's important because some index AMs
* expect a relcache flush to occur after REINDEX.
That is, what we need to force an update of is either the relcache's
rd_indexlist list (for a table) or rd_amcache (for an index).
In the REINDEX case, we could conceivably skip the flush on the table,
but not on the index. I don't think it's worth worrying about though,
because REINDEX will very probably have an update for the table's
physical size data (relpages and/or reltuples), so that it's unlikely
that the no-change path would be taken anyway.
regards, tom lane
