On Tue, Aug 18, 2009 at 03:01, Andrey Kuzmin<andrey.v.kuzmin at gmail.com> wrote: > The patch itself seems to be "purely opportunistic" :), hence two questions: > 1. How does it impact prefetch efficiency?
On my workloads it does not. Whereas lock contention with multiple threads accessing a file is quite high. > 2. If it does impact prefetch significantly, is there any other, less > destructive, locking approach? As the comment in the file indicates one could make the locking finer grained. -Kip