The Page Cache

To avoid unnecessary disk accesses, the kernel never tries to read a page from disk without looking into the page cache and verifying that it does not already include the requested data. To take the maximum advantage from the page cache, searching into it should be a very fast operation.

The unit of information kept in the page cache is, of course, a whole page of data. A page does not necessarily contain physically adjacent disk blocks, so it cannot be identified by a device number and a block number. Instead, a page in the page cache is identified by an address of a special data structure named address_space, and by an offset within the file (or whatever) referenced by the address_space data structure.

Continue reading here: The addressspace Object

Was this article helpful?

0 0