On Thu, Mar 17, 2011 at 08:46:23AM -0700, Curt Wohlgemuth wrote:

But if one of one's goals is to provide some sort of disk isolation based on
cgroup parameters, than having at most one stream of write requests
effectively neuters the IO scheduler.

If you use any kind of buffered I/O you already fail in that respect.
Writeback from balance_dirty_page really is just the wort case right now
with more I/O supposed to be handled by the background threads.  So if
you want to implement isolation properly you need to track the
originator of the I/O between the copy to the pagecache and actual
writeback.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>