> NOTE: I'm saying that it is _bad_ to throw away readahead pages
> easily because I've seen it in practice. I'll try to test this
> in real practice again soon (not this week) to make sure.
>
> Comments ?
Setting the referenced bit on not (yet) used readahead
pages could have the damaging side effect of swapping
out pages from the working set in preference to the
readahead pages.
OTOH, without some special trick for linear access the
VM will prefer to swap out the pages we're about to use
(the readahead pages) instead of the pages we just used
-- the absolute worst thing for sequential IO.
One simple solution which achieves both the effect of
dropping the right pages in a sequential IO stream and
making sure that not-used readahead pages aren't pushing
out the working set would be drop-behind.
I know drop-behind isn't the prettiest thing, but it's
simple and it works.
regards,
Rik
-- IA64: a worthy successor to the i860.http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/