Hi...
Does the page cache keep pages that are discarded but anticipated to get reused by processes, or does it actually maintain pages that are used and already within the address space of processes? nnnnnn
To the best I know: once you read a portion of file, its pages enter the page cache. So no matter the status is (used or not used by a process), they will stay in page cache until kernel VM subsystem decides to reclaim them in order to provide more free pages.
IMHO, the goal of page cache mechanism is for optimization. For tracking used pages, you can't really rely on just checking page cache entry, since they don't have to belong to certain process address space. Maybe reverse mapping can better serve this purpose.
regards, Mulyadi -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
