You still have a buffer cache, but Linux says it is for meta data only.
Has the limits (Default) been adjusted for such?
Finally, Linus even says that things haven't been optimized a great deal
yet. Here is a question I really must ask:
If the normal use for page cache and what used to be most of the buffer
cache now share the same space... do they still have different limits,
or are the old uses and new uses of the page cache forcing each other
out of the cache?
Honestly, I have only reviewed some of the code, and I only understand
about 50% to 80% of what I read... I could be totally off whack here,
but it seems to me that this could very well be what is causing the slow
down, you have the two uses fighting for a overall "smaller" cache...
causing all sorts of slow downs as different "needed" pages get bumped
out.
Sincerely, in guess work only,
Trever Adams
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/