Re: Bug in new buffer freelist code

Jan Wieck <JanWieck(at)Yahoo(dot)com> writes:
> It seems to me that buffers that are thrown away via
> StrategyInvalidateBuffer() do not get their relnode and blocknum cleaned
> out.

Mmmm. They definitely should be; if you look at the prior version of
buf_table.c, BufTableDelete did this:

/*
* Clear the buffer's tag. This doesn't matter for the hash table,
* since the buffer is already removed from it, but it ensures that
* sequential searches through the buffer table won't think the buffer
* is still valid for its old page.
*/
buf->tag.rnode.relNode = InvalidOid;
buf->tag.rnode.tblNode = InvalidOid;

I see you removed that from the current version, but the equivalent
thing needs to be done someplace. Is StrategyInvalidateBuffer the
right place?

BTW, it kinda looks like the BM_DELETED flag bit is useless now?

regards, tom lane