>I would ask Chuck to bench this my new code (it should be an obvious
>improvement).
Don't try it!! I have a bug in the code that is corrupting the cache
(noticed now).
I am working on a fix now.
I just noticed that kmem_cache_reap can theorically sleep in down() so
here I moved the cookie++ in do_try_to_free_pages(), it will be less
finegrined but safer... But I still have corruption so I am looking at
filemap.c now...
BTW, why kmem_cache_reap() need a serializing semaphore?
Andrea Arcangeli
-
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/