FTS, LRU/MRU and X$BH

I'm doing some IO testing on new hardware from an Oracle 10.1.0.5.0 on AIX. One of my tests involves FTSing a 45GB table with our buffer cache set to 3.3GB. My testing with done with minimal load on the test database with only idle ERP connections, lightweight scheduling software, and GC Agent.

I had flushed the buffer cache and confirmed with a query on X$BH. After running the test query, I examined X$BH again, but saw no blocks from my test table. I confirmed that the docs at http://tinyurl.com/a6n4bz say FTSing will cause blocks to be put on the LRU end of the buffer cache, so I expected to at least see them at some status. Here's the query I used:

SELECT so.name object_name, COUNT(1) buffers FROM v$bh xbh, sys.obj$ so
WHERE so.obj# = xbh.objd
GROUP BY name
ORDER BY 2 DESC; I'm thinking I've killed some 13yo Oracle Education brain cells somewhere, but as my single table test is repeatable with almost all physical reads each time, it seems like the blocks are never hitting the buffer cache, which can't be true. This doesn't seem to jive with the above docs or my limited understanding of the x$bh view.

Thoughts? TIA!

Rich

--
http://www.freelists.org/webpage/oracle-l